Video crash at least once a day...

[rfatcheric]

At random times while gaming I will have a video crash. Current game locks up, sometimes I can alt+tab out or alt+F4 to end the game but then im at my desktop at a very low resolution with like 256 color scheme, almost like a driver crash or something. I get no error that pops up and nothing logged in the event log for my system. I have to reboot for the problem to be corrected.

This can happen in 20 minutes or 2 hours, it seems to be completely random.

I dont seem to be having any heat issues.

My system is
a64 4200+ x2 @ stock speeds. Runs about 29c idle and 39-42c @ load
8800 GTS 320 @ stock speeds. Runs about 45c idle and 54-57 @ load
1bg Mushkin value ram (2x512)
Asus A8N-SLI mobo
Not at home so Im not sure about the PSU but I know its an Antec PSU


Depending on the game that im playing, the video will either freeze completely but I can still hear sound, and the game still reacts to my mouse / KB commands, or I will get severe tearing and artifacting followed by video freeze.

At first I thought it was a heat issue but I ran a log so I could see what temps looked like while gaming and temps are fine.

I have reformatted and the problem still persists. The problem is most notable when I play World of Warcraft. It seems to happen a lot more often in that game, but the problem definately isn't limited to WoW only.

Any thoughts?

** edit ** forgot to mention, im using the latest drivers for all hardware in my computer. I am running WinXP pro sp2

 

[tomdrum]

im not really sure....This has happened to me (8800GT) while playing a game that isnt very graphics intensive....I can play crysis fine at high but while playing project tourque it sometiems decides to artifact and freeze the whole system. Im not sure whether its the heat

Im thinking PSU

 

[mactronix]

im not really sure....This has happened to me (8800GT) while playing a game that isnt very graphics intensive....I can play crysis fine at high but while playing project tourque it sometiems decides to artifact and freeze the whole system. Im not sure whether its the heat

Im thinking PSU

Two main culprits for my money would be either Bad PSU or the card is slowly dieing on you.
Mactronix

 

[ggorrell]

I have had the same problem with my system since I bought it (mainly in WoW as well, but also other games, a couple of times while just running the screen saver). It will lock up without warning. Sometimes it will unfreeze after a few seconds, sometimes the sound will come back, but the video is frozen, and sometimes it will simply reboot itself.

In trying to diagnose my problem I have been through 3 video cards (Geforce 6600, 6800, and now 7600GT), 3 power supplies (the original fried and was replaced, I then bought another brand hoping it was a problem with that brand), and 2 sets of memory (2 sets of 1M each, which I now use together). None of which fixed the problem. The only thing that will work is pulling the video card and running off of motherboard graphics (which are horrible).

I have given up on fixing the problem (I am convinced it is a motherboard problem probably related to the video card socket), and am hoping to just replace the whole system in another month of two.

 

[rfatcheric]

Two main culprits for my money would be either Bad PSU or the card is slowly dieing on you.
Mactronix

I hope the card isn't dying

Previous to this card I had a 7900gt that had almost the same symptoms and eventualy it got to the point where the 7900gt would do it in seconds rather than minutes or hours so I stopped gaming on it and then I started having problems even in windows browsing the net and such, I had random colored single pixel dots on the screen that even showed while in DOS mode so I tried the 7900gt on another system and it had the same random colored dots so I figured it just went up and replaced it with my 8800gts 320.

The 320 didnt seem to have this issue for a long time, just started a few weeks ago, I've had it for almost a year.

Now, having used 2 different cards now with both getting almost the same problems I would think it is something else in my system. At this point I think I'll just suffer and wait till I upgrade next year. I plan on getting a new mobo/cpu/ram/psu eventually. I'll see how this card works out on the new system... if its still alive by then.

Thanks for the replies, I think right now its just going to have to remain a mystery. I dont have another system to test in, and I dont have integrated graphics or another PCI-E video card to try and figure out what the problem is.

 

[leon2006]

Sounds like a memory related problem. Have run memtest in your pc?

It won't affect you until you use the that bad location.

You may consider adding memory capacity. 2x512 is pretty small specially for games like crysis.

 

[tomdrum]

Well for my situation im a bit confused, as every other game i run has ben fine except fro the one (project torque). My card is only a few months old (OC 8800gt with zalman) and doesnt crash on me (but isnt playable) when attempting to play crysis on high with 4xaa, yet this game is pretty basic when compared.

The exact problem that i get is that alot of little grey squares appear occasionally, the sound loops on a particular noise and then it just freezes like that. Odd for me

 

[leon2006]

Have you tried to use other PCIEx16? If its HW it can be on the following:

1) Mobo
2) Memory.. If its not a HARD ERROR it can be a SOFT ERROR. Both can crash application.
3) Video Card: Video artifacts are signs of bad memory on the card
4) PSU:....Measure your voltages at the connector when loaded. Make sure the voltages are with in specifications. What is the capacity of your PS?

 

[gomerpile]

These issues sounds like 2 things. Timing set 0 of the vram has lost its timing or the cache memory has an error. I'm more to lean on the Vram because usually with the cache memory there would be a pagefault error.
You can check your video card by using nibitor to read your bios try to change the timing set 0 then test, if this receives and error thats the problem.
One problem is the voltage table when the video receives an error with some brands of cards, the video card reduces the power voltage table to table 2 or 3 and it does not change back.
you can fix this with nibitor by changing the 2,3 voltage table to table 1.

 

[Kkkk1]

After reading the above some great replies but if it was me i'd start at the basics.

I'd try installing your card in a freinds PC and see if you experience the same. If you do at least you know the video card is dieing. If it doesn't i'd look at the PSU next. Try instlling the same friends PSU in your PC and try the same test. At least you can narrow it down before you go replacing components. It definately sounds like a hardware issue somewhere.

 

[rfatcheric]

After a few days of testing I have only found that when running in "windowed" mode this problem does not occur.

Also, after disabling the auto restart on error option in winXP I do get a blue screen sometimes when this problem happens, other times I dont.

The bluescreen is an error with nv4_disp and the error is something to the effect of 0x000000EA

But it doesn't happen while in windowed mode.

 

[mactronix]

After a few days of testing I have only found that when running in "windowed" mode this problem does not occur.

Also, after disabling the auto restart on error option in winXP I do get a blue screen sometimes when this problem happens, other times I dont.

The bluescreen is an error with nv4_disp and the error is something to the effect of 0x000000EA

But it doesn't happen while in windowed mode.

This may be usefull http://support.microsoft.com/default.aspx?scid=kb;EN-US;q293078 also some with the same message reported that they ran memtest on the ram and found it to be faulty,replaced ram and problem went away.
Running in windowed mode would make life easier for the card which brings us back to the card having issues or the PSU not giving it enough juice, You said it was an ANtec can you tell us model number and wattage ?
Mactronix