Hardware Canucks

Hardware Canucks (http://www.hardwarecanucks.com/forum/)
-   Troubleshooting (http://www.hardwarecanucks.com/forum/troubleshooting/)
-   -   textures changing in game, freezing - video card problem? (http://www.hardwarecanucks.com/forum/troubleshooting/11448-textures-changing-game-freezing-video-card-problem.html)

tsincaat November 2, 2008 04:26 PM

textures changing in game, freezing - video card problem?
 
I've had these problems for quite a while now, but I'm finally posting about this because I can't stand it anymore.

I have an EVGA 8800GTX.

This seems to be an off and on sort of thing, though as I sometimes don't play a whole lot of games the problem may not be intermittent at all.

I don't know the correct term for this. Basically, once in a while, I'll be playing a game, and everything will freeze up, then, all the textures on everything changes and turns to slo-mo where i'm lucky to get a single frame every 5 seconds.

I took this screenshot a while ago with warhammer:aor:
http://i179.photobucket.com/albums/w...rdproblems.jpg

As far as I can recall, the sound always continues to play in the background, and usually, if I wait long enough, I can alt tab out to task manager.

The last games I can remember having this problem in have been Warhammer:aor, Age of Conan, the Call of Duty 5 WaW beta, and just a few minutes ago, Fallout 3, though there have been others in the past.

I'd love for someone to tell me what's going on :]

DarKStar November 2, 2008 05:09 PM

Are you overclocking your card and/or CPU ? Try running everything at stock speeds. If not OCing, have you checked your card's temperature to rule out overheating? What kind of PSU do you have ? If your PSU is good enough and it's not a CPU/Overclocking/overheating issue then I'd check your RAM, otherwise (memtest86) - and after trying stock settings, check temps, checking RAM, PSU, then I guess your card is bad.

tsincaat November 2, 2008 05:26 PM

Quote:

Originally Posted by DarKStar (Post 107576)
Are you overclocking your card and/or CPU ? Try running everything at stock speeds. If not OCing, have you checked your card's temperature to rule out overheating? What kind of PSU do you have ? If your PSU is good enough and it's not a CPU/Overclocking/overheating issue then I'd check your RAM, otherwise (memtest86) - and after trying stock settings, check temps, checking RAM, PSU, then I guess your card is bad.

Thank you for the response Dark.

I'm not overclocking.

Opening nvmonitor from the nvidia control panel it says "nforce" (I assume my motherboard) is at 57 c and my video card at 62 c (no games running).

I think my PSU is an 850w, when I purchased it I was sure to choose a decent brand. However, I can't remember the name or find the box around, and as the sticker isn't on the back/top/side (viewable parts). It might be on the bottom, but I'd have to take it out to check, and I'd rather not bother unless the brand is important info.

Edit: I ran memtest and it said that no problems were detected.

DarKStar November 3, 2008 10:19 AM

Something is not right with those temps. There is no way your motherboard can be running at 57 C - unless you have SERIOUS heating issues in your case or the sensor being close to your GFX or misreported. I use EVEREST to monitor my temps. Also 62C idle is too high.

Try getting GPUCaps Viewer and run a Fur Test, with max settings, while monitoring your GFX card temp - if it keeps going up and getting ready to reach 100C stop it right away.
Look for any artifacting or freezing during the testing as well.

Try to find a beter utility and report back with your real motherboard, CPU, and gfx card temps.

GPUCaps here
GPU Caps Viewer: graphics card information utility, GPU temperature and OpenGL API support | oZone3D.Net

Remember when running these tests always keep an eye on the temp and be ready to stop the test as needed -

You might want to try SpeedFAN, it's free, although depending on your setup it might misreport some voltages and some info.

tsincaat November 3, 2008 11:38 AM

Thank you for the reply dark:

Everest reports my GPU temperature idle at 54c
GPU Caps reports my GPU temperature idle at 61c

Running the test with 32x AA and 768 x 1200 on a 768 x 1366 32' LCD HDTV

80-81c with Everest
85-86c with GPU Caps

Interestingly no artifacts or freezing.. I'm kinda wondering if it's somehow related to software or drivers? Last night when I was having problems with the screen going black and freezing on the desktop a restart seemed to greatly help and I haven't been encountering that.

In terms of Motherboard and Case:

I have an Asus crosshair.
I have an antec 1200 case with the front and rear fans at 800rpm (any higher and I get annoyed)

Thank you for all your help so far.

Squeetard November 3, 2008 11:43 AM

What are your cpu temps under load?

tsincaat November 3, 2008 12:13 PM

Everest says the CPU is at 45-46, Core 1 at 45-46, and Core 2 is at 40

AMD 6000+ 3.0 GHZ dual core

DarKStar November 3, 2008 01:04 PM

Ok your temps are fine.... So are you saying that GPUCaps your temp under full load does not exceed 86C? you are fine then. that's ok - On one of those faulty, overheating chips temps would go much higher.

CPU temps are ok (LOL - is that 6000 on stock HSF ??? because you are running way cooler than I am with my X2 5600+ stock HSF :D)

Definately not temperature related or overclocking. WHat about RAM, brand/Model and timings if you know them .... Are you running your modules according to their specs, is the voltage set correctly - if you are using agressive timings, try bringing your RAM back to relaxed settings, try 5-5-5-18 just for testing purposes, and tell me if you are still having problems. If it does the same thing, then you ruled out RAM, CPU, temperature, and PSU, well actually if you can provide brand and PSU ratings in amps for each rail, might help....... Try uninstalling your drivers - then booting in safe mode and using DriverSweeper from guru3ds site and completly remove traces of NVIDIA drivers - then re-install the most recent set of drivers again. If problem persists then you might have faulty RAM on your graphic card itself or a bad card and the failure happens to be at the far end of the RAM address so it shows in games with big textures - I wouldn't rule out the possibility of a bad card - assuming that everything else was checked first (CPU, RAM, proper settings, etc.)


All times are GMT -7. The time now is 09:14 AM.