Graphics card problem? Or something else?

[Fromez]

Right, a few weeks ago I got myself a new mouse to replace a wireless one, which was getting annoying to replace the batteries. Since then I've been having intermittent freezing of the cursor where when I waggle it it stops moving for a second. Up til now, I thought this was the mouse and made a mental note to replace it, yet again.

Yesterday, I played Tomb Raider Anniversary for a bit and started getting graphical artifacts (blockiness, fuzziness). Tried again, the game wouldn't start and constantly got display driver stopped errors.

Rebooted, tried again and the game crashed in-game with a bsod (my first on this system). Don't have the software to read the minidump but the Vista help screen said something about hardware fault.

Reinstalled older nVidia 186.18 drivers, the freezing seems less frequent.

Tomb Raider Anniversary works fine now for the time being, but still got the freezes. Noticed on Medal of Honor Airborne, there are green pixels on the black loadout screen when I'm quitting the game, but nowhere else. Crysis runs fine on medium, but the intermittent freezing is apparent in-game.

Motherboard and CPU temperature is fine, downloaded nVidia tools to see if they have a temp gauge. Reseated gpu. Reseated memory.

Funnily enough, my old Pentium 4 system (that lasted me 5-6 years) started developing identical graphic artifacts with...Tomb Raider Legend. In the end it was pretty obvious the gpu had given up as the whole OS would boot up in 4 colour mode.

But I have an entirely new system, processor, motherboard, memory, gpu, new and higher wattage PSU etc.

It's coming up to 362 days since I received this new graphics card (and built this system) and so I want to determine whether this is a hardware fault. But I don't want to send my card back and they find no errors, since I will be without access to this system for a longish period.

Is there anything else I can do to find out?

If the minidump from the Tomb Raider Anniversary crash'd help shed light on the issue, I have saved it if anyone knows how to interpret it and what I need to install to read it. Never needed to do a minidump in my life as I've only had these 2 major problems ever.

Thanks for your time and reading this longish post!

 

[Jason1973]

Hi Fromez,

From what you describe it sounds to me like a heat issue. My friend was suffering the same graphical problems.

Download and run furmark in stress test mode.

If it is heat related you will see the stutter onscreen and the temp drop a few degrees before recovering.

You might get away with a blow out to clear obstruction or it might require an RMA.

Good luck and post back with the results

 

[Fromez]

Did a Furmark benchmark. Ran for 60 seconds fine, score 1114.

Did a stability test, 1024x768 (my resolution), 0MSAA, no post processing, started off at 47 degrees, was no stutter, ran it for 240 seconds, from about 120 seconds to then it was a stable temperature of 73 degrees and no stutter or crashing.

 

[Fromez]

Sorted the mouse freezing problem (was the mouse - took it back)

Stil have the nvlddmkm "display driver has stopped working and has recovered" problem though. Done a Vista reinstall twice and still consistently get a crash running Crysis in DX10 mode (works with DX9 for some reason). Looked at temperatures but they're all fine. Done a memory test and that's fine.

I read on a different forum that it might be to do with ampage. Can someone look at these specs of my PSU and tell me what the ampage is?

Edit: is it 17A?

Here's the snippet from the post I read:

my old psu was runing at 18 amps for +12 volts. pretty pathetica top be honest. anyway my new psu runs at 56 for +12v...normally as soon as i start a graphic intensive game it crashes straight away... even on low/medium settings. so i thought id go all the way and tried one of my games on everything maxed out to the super maximum. including AA etc... no crashes whatsoever.