Unstable system

[Zachyattacky]

Wow I love this new layout... Anyway, I recently put together my new computer and it started up fine, drivers did fine etc etc, but now this week I started pushing it a little bit harder with more demanding games, higher fps in others etc.

At the start of the week I started getting driver errors, so I lowered the fps, and I never had a problem. But as soon as I raise the fps a bit after awhile it it crashes.

I tried crysis. On dx10 it would get to the load screen of the first mission, finish loading and go black, dxerror. I tried it in dx9, I could play on ultra high silky smooth, after about 5 min of play, black screen, dxerror

So last night I redid the drivers and directx, and the same thing still happens, I tried furmark and after about 30-60 seconds on higher resolutions it crashes and says furmark stopped working (when I do the gpu temp burn it crashes before the gpu gets over 70)
My cpu temp I've never seen over 34.

System specs are:
x58 classified
gtx295 ftw x2
12 gigs ddr3
i7 920 @ 3.19

So...what do I try next :S

 

[meat_helmet]

Are you ever getting BSODs?

What program are you using to monitor temps? 70C actually seems very low for a 295 considering my 285 will go up to 85C with no problems. 34C for an i7 also seems too low for load temps unles you are running water. Check temps with Real Temp.

 

[Zachyattacky]

Hey, I'm using the evga tools (precision and the eleet) for my temps, and the 34 for my cpu isn't at full load, that was just after an hour of cod4 (and I'm using cooler masters v8 rifle to cool it) and no, I never get any bsod's, it either says the program stopped working, the drivers have failed or there was a directx error.

Thanks for the reply

 

[Zachyattacky]

Okay as of today I'm stumped... I tried each of my cards by itself, and ran a stability test and benchmark with Furmark, they both flew through the stability test flawlessly (15 min of it), and they got the exact same benchmark...the only difference was one card got as hot as 76c and the other only was running at anywhere from 42-44c, I did the xtreme burning mode both times and the exact same resolution... as soon as I put the cards back in together it crashes after 34-91 seconds, and the NVLDDMKM driver error pops back up...

Just for giggles I made sure each stick of ram was good, and it all checks out... so it has to be something software related, but I have no idea what to do next....

 

[Zachyattacky]

Okay so...sorry for post spamming but aparently the factory oc was unstable because I just underclocked my timings to 610/1310/1010 and now the stability test is still running fine after 5 minutes...so I guess that was it... Strange that a factory oc wasn't stable... :S

 

[Slaymate]

Contact the manufacturer. Most reliable companies will RMA a card with heat issues.