Unknown Hardware Issue

[Celcius]

I've been trying to diagnose an issue with my system for about 3 months now lol. I'll be playing a game and then all of a sudden my screen will go black or another solid color and even though all the system fans are still running I have to manually reset the machine. No blue screen code or other error message upon reaching windows, but I can go to windows event viewer and see event ID 41.

So far I've tried:
Re-installed windows
Different videocard (when I used a gtx 680 I got a blue screen code 116 message upon going back into windows, but with the gtx 280 I get no error messages)
Different RAM (upgraded from 8gb to 16gb just because I wanted, but the issue still remains)
Different power supply (also tested an AX850 psu)
Added more fans to case
Different videocard drivers

and my current config is:
Intel 2600k @ 4.5ghz
Noctua NH-D14
Asus P8Z68V-Pro GEN3
16GB G.Skill Ripjaws X
Intel 160GB 320 Series ssd
EVGA GTX 280 SC
Seasonic X750
Corsair 550D
X-Fi Titanium HD
Samsung dvd burner

I can run prime 95 and other non-3d programs with no problems. Maybe it's possible that both videocards I've tested were bad, but unlikely. The most frustrating part is that the issue may happen after 3 minutes or 3 weeks, and I've only had it happen while playing a game or running a 3d stress test (though I ran the heaven bench in loop all night long with no problems).

Any other ideas? Perhaps the pci-e slot or mobo going bad?

 

[ob1]

Have you tried to drop your O/C down to like 4.2 for a week or so and see if that changes the symptoms?

 

[Celcius]

I thought I remembered having it happen even at stock, but I'll see if it happens again. Earlier this year I logged 100+ hrs gaming with no issues (same overclock), but since replacing the psu and videocard in May I've had this issue. The strange thing is, I've tried replacing both and can't seem to get rid of the problem...

 

[/usr/sbin]

I know you tried different video cards, but did you try the other PCI-E slot?

 

[Celcius]

I know you tried different video cards, but did you try the other PCI-E slot?

I've thought about trying that....First I'll see if the issue happens with the cpu at stock, then if that fails I'll try the other slot. I just wish it happened more frequently so I didn't have to wait so long between tests. Thanks for the suggestions guys

 

[jahara]

I thought I remembered having it happen even at stock, but I'll see if it happens again. Earlier this year I logged 100+ hrs gaming with no issues (same overclock), but since replacing the psu and videocard in May I've had this issue. The strange thing is, I've tried replacing both and can't seem to get rid of the problem...

Why did you replace the PSU? I mean its technically possible that the current PSU isn't as stable as the past one which results in worse OC's.

 

[Celcius]

A few days ago I bought a Corsair AX 850 and yesterday the issue happened again which makes me think that the psu isn't the issue. Originally I had a corsair hx 850 but only changed to the seasonic because I had never owned one of their units and liked the quieter fan, full modularity, and 80+ gold certification. (plus it was on sale lol)

 

[Celcius]

I'm still testing gaming at stock, but I've noticed that when I try to run prime95 on fft size 1344k, whether my cpu is at stock or overclocked, to stop the test from running I have to manually kill the process (never any errors though). No other fft size does this. Is this a program bug? I'm using prime 95 version 25.8.

edit: nevermind, I've downloaded the newest version of prime and the issue is fixed. Now back to testing games...

 

[Celcius]

Well, I finally found the problem - my overclock has become unstable. Now that I have the latest version of prime, I've started back stress testing and I get blue screen 101, which means it needs more vcore. I guess 1.3v is no longer enough...