Bought a new computer a couple of months ago, and all has been fine and dandy.

Until a couple of weeks ago that is. I started to get sudden restarts while playing demanding games. Over time however, the amount of demand on the computer to force a restart has gone down and down.

I havn't been able to tie the issue to anything, seeing as the restarts seems almost random, aside from the above mentioned.

Watching the event viewer, the error message I always get is Event 41 - Kernel Power issue. Taking a closer look at the error message tells me nothing, basically a timer on how fast the computer reboots at the different stages.

These are the critical errors avaliable in the event viewer and event viewer for performance issues respectively. As far as I've figured they're simply descriptions of the fact that the computer suddenly restarted, and not tied into the issue itself.

The thing I suspect is happening is power related, some part simply isn't getting enough. I've reopened the case and switched out the power cables, didn't fix the issue. Thing is, the problem started whilst I was replaying SWTOR, and it's not the first time I've done that with a new computer, not to mention as I said earlier, the issue seems to be working itself down.

Any help or insight would be appreciated.

 

This means that the System didn't have enough time to create a STOP error (mostly known as Blue Screen). Since that normally happens if your PSU is defect or the PC overheats, I would check those two things first.

 

800w on the PSU. 8gb RAM. 4gb avaliable, 2gb dedicated and 2 gb shared.
Videocard = Geforce GTX 670. I've tried reformatting, and I don't have another computer avaliable, only a laptop. Google of the error codes (Kernel 41 etc) seems to be one of the most speculated and never solved issues on the internet. People gets the error in a plethora of ways with none to very few solutions.

I've been 'monitoring' the gpu temperature whilst playing and while I see an increase (30-35c -> 60-65c (now, with only internet and standard programs running the temperature is down to 27 degrees) while playing demanding games, it's not overheating to the degree that it should cause problems. PSU would be a probable cause, but I'm at loss at how to actually pinpoint the problem specifically to it, I'd rather not waste time sending it back just to get a note that it's not the actual problem. Any tips on how to check it myself?

 

If possible, try to get your hands on another PSU to test if your PC still reboots with it. If yes, you know that it isn't the PSU and if no, well you can send your PSU in. That would be the easiest way to test it.

As for the PSU itself, you could use a multimeter or PSU tester to look if all the voltages are normal, but I don't know if that would really help, since your problem only occurs if your GPU is pulling the 170W that it needs. You could go to the nearest PC workshop and ask them to test it, since they should have the necessary equipment, but that's probably not exactly cheap.

 

A 1200W power supply is absolute madness for a single graphics card (a full system with a high end graphics card will use ~400W maximum). A good quality power supply is what the OP should be looking at.

 

Speculation that it's a failing PSU or overheating CPU or GPU is probably correct.

It could also be a not quite seated properly plug or component. Heat rises which increases resistance which causes the borderline functioning connection to fail. I've had cables in systems which haven't been moved in years come unplugged, so don't rule it out without checking.

Use stress testing software to stress just the CPU, then the GPU, the RAM, etc.

If GPU test fails: Most modern CPUs (on AMD) and Mobos (on Intel) have onboard graphics. Switch to the onboard (via the Bios) and stress test the graphics again. If it passes, then that's confirmation.

 

Checked temperature across the board, motherboard, gpu, cpu, you name it. All coming across as very fine. Checked performance of graphic card, nothing stands out.
All plugs, wires and components are as tight as they ever was.

The random restarts still occur.

Pretty much narrowed it down to a somehow failing PSU, I'll send it in for a check during the holidays when I vacation or are otherwise engaged anyhow.

Thanks for the help.