Another 6800GT bites the dust?

[Goi]

I bought my XFX 6800GT around late august last year, and it has been performing great. However, recently, when I got back from winter break, I discovered that my 3D games such as Q3 and Doom3 would exhibit gradually worse artifacting within minutes, and if I don't exit the game immediately, the system would either freeze, show a BSOD with memory dump, or reboot.

The system is currently running stock(no overclocking in any component), so that shouldn't be a problem.

Does anyone have any clue what's wrong? I sent an email to XFX, hopefully they'll reply. I've heard bad things about them though.

Thanks.

 

[VIAN]

The problem is XFX. Good luck with their customer service. Next time, don't cheap out and get a BFG or Leadtek.

 

[Goi]

Well, the cards are all reference boards so I don't think it's that much of a problem. Their customer service might suck, but I don't think the cards themselves can be much different if they stick to reference designs, which I think they have, since even the HSF sticker is Nalu.

Any other more constructive opinions/suggestions?

 

[302efi]

Sounds like a heat problem..whats the temps of the card ?

 

[Goi]

That's what I thought too, but after exiting from the games, the temps are in the high 60s according to the control panel, which seem normal.

I've tried downclocking the card to 300/800, and still this problem occurs.

 

[Cdubneeddeal]

Could it be possible that there's not enough thermal paste on the chips?

 

[The Green Bean]

Originally posted by: Goi
That's what I thought too, but after exiting from the games, the temps are in the high 60s according to the control panel, which seem normal.

I've tried downclocking the card to 300/800, and still this problem occurs.

high sixties seems a bit high when exiting games. Run Riva Tuner's hardware monitoring. I think you in game temps will be 90+...But since the operating NVidia temps seem to be upto 135C, You are entitled to an RMA.

 

[Gamingphreek]

Originally posted by: Goi
I bought my XFX 6800GT around late august last year, and it has been performing great. However, recently, when I got back from winter break, I discovered that my 3D games such as Q3 and Doom3 would exhibit gradually worse artifacting within minutes, and if I don't exit the game immediately, the system would either freeze, show a BSOD with memory dump, or reboot.

The system is currently running stock(no overclocking in any component), so that shouldn't be a problem.

Does anyone have any clue what's wrong? I sent an email to XFX, hopefully they'll reply. I've heard bad things about them though.

Thanks.

Not to start anything but your rig says (415/1150).

-Kevin

 

[genghislegacy]

dont have a clue. did you install anything after the winter break? also might try some different drivers.

 

[302efi]

Originally posted by: Gamingphreek

Not to start anything but your rig says (415/1150).

-Kevin

Yeah you trying to start somerhing by posting the bait so others can bite on it..you could have PM'ed about it.

Don't turn this into another geforcetony thread

 

[Mrvile]

Video cards often cool down very quickly. Run rthdribl while monitoring your temps at the same time to see where they go.

 

[Gamingphreek]

Originally posted by: 302efi

Originally posted by: Gamingphreek

Not to start anything but your rig says (415/1150).

-Kevin

Yeah you trying to start somerhing by posting the bait so others can bite on it..you could have PM'ed about it.

Don't turn this into another geforcetony thread

I didn't turn it into a GeforceTony thread in the first place. I was very polite that entire thread until the OP starting flaming, and blatantly lying.

Dont try to pin this crap on me. I simply stated what was in his sig. His video card was OCed and is failing. Im not saying anything else or instigating anything.

-Kevin

 

[302efi]

Originally posted by: Gamingphreek

... His video card was OCed and is failing...

... Im not saying anything else or instigating anything.

-Kevin

That sir, is a oxymoron:laugh:

Why didnt you PM about his sig before posting it ??..You been around here awhile and you know what happens when somebodys video card is OC'ed (not say the OP's card is) and failed

 

[bjc112]

Originally posted by: Gamingphreek

Originally posted by: 302efi

Originally posted by: Gamingphreek

Not to start anything but your rig says (415/1150).

-Kevin

Yeah you trying to start somerhing by posting the bait so others can bite on it..you could have PM'ed about it.

Don't turn this into another geforcetony thread

I didn't turn it into a GeforceTony thread in the first place. I was very polite that entire thread until the OP starting flaming, and blatantly lying.

Dont try to pin this crap on me. I simply stated what was in his sig. His video card was OCed and is failing. Im not saying anything else or instigating anything.

-Kevin

Just because his sig has it listed as overclocked doesn't mean it currently is.. Especially when he says every componet is running @ STOCK.

I highly doubt he is lying.

 

[ponyo]

Gamingphreek = Rollo Jr.

 

[orangat]

Then he should change his sig (or explicitly state that it only tested for oc at that speed).

 

[Gamingphreek]

I simply asked a question. Im not trying to cause problems. If your sig is wrong or was entered wrong then just remember to change it so people dont get the wrong idea.

-Kevin

Edit: Goi is a long resident and rspectable forum member. Im sure everything is fine, i was just wondering.

 

[Goi]

The rig was previously overclocked, but has been running stock since I got back from winter break. Sorry for the confusion As part of the troubleshooting process, I also downclocked various components to test if this was a problem. That's why I said the system is currently running stock.

 

[Goi]

I ran rthdribl and monitored my temps. According to the 6800GT control panel, the GPU idle temps are around 56-57C, with an ambient temp of 38-39C, and the max temps I got before serious video corruption on rthdribl was 78C/44C. I ran this twice and got similar results, then after the 2nd run, I began to get CPU usage spikes every few seconds, which rendered my system unusable until a reboot. I tried to track which process was causing the spike, but the time slices of task manager processes window wasn't small enough for me to capture the offending process. The CPU usage history graph does show the spike occuring every few seconds though. I let the GPU cool down for a few minutes back to around 58C again, but the spikes were still occuring, so I had to reboot.

 

[Gamingphreek]

Originally posted by: Goi
The rig was previously overclocked, but has been running stock since I got back from winter break. Sorry for the confusion As part of the troubleshooting process, I also downclocked various components to test if this was a problem. That's why I said the system is currently running stock.

That explains a lot !

Well as for the problem, what PSU are you using? YOu might be having voltage spikes, which is increasing and decreasing the clock. Try different molex connectors.

Also if your memory is OCed at all bring it back to stock because if you get that stop error something is wrong in the caching process.

-Kevin

 

[Goi]

The PSU is an Enermax Noisetaker 470W, which I believe to be sufficient. The problem doesn't seem random, but seems to occur more frequently as time passes. Because of this, I'm inclined to think that it's a heat problem.

Like I said, currently the entire system is running stock. In fact, I even tried to bring the graphics card all the way down from the stock 350/1000 to 200/500, and still this happens. It certainly took a while longer, but while running rthdribl, the temps went up to around 72-73C this time(took a lot longer to get there than before) before the system crashed with a BSOD/memory dump on nv4_disp.dll, which certainly indicates a graphics card problem.

I also tried disabling Fast Writes and Sideband Addressing, and after a while of fragging in Q3, the problem occured again.