Computer randomly crashes.

[keithf]

Hi there guys and gals first time posting.

I have a ~6 month old pc that I built after the south bridge on my old motherboard got fried. The only parts that I transferred over to my new machine were my power supply, keyboard, mouse and headset.

The problem is anytime I am gaming (3d models only, able to play terraria and emulator games fine, and have never crashed at desktop), my computer will crash and reboot. It sounds like my video cards are getting hot, but ccc says they are around 40c under load ( is that too hot?).

I have tried two different video cards. A msi twin frozr 560ti and y current card set up of 2 xfx 1gb 6870's running in crossfire. I have tried the 6870s individually and get the same result.

I have tried two different sets of ram a 8gb kit and two 8gb sticks together.

Motherboard: ASUS p8p67 deluxe rev 3.0
Processor: I5 2500k sandy bridge
Video card: 2x xfx ati 6870's
Ram: 2x 8gb Corsair vengeance (cmz8gx3m1a1600c10)
Hd: Ocz vertex 3 120 gb ssd ( have a independant 1 tb seagate barracuda on backup)
Power Supply: Thermaltake toughpower 750w (possible culprit?)
Os: Windows 7 64 bit

If anyone has any information it would be appreciated, and I'm more than happy to provide more information.

 

[jimmysmitty]

40c is not too hot but under load seems too cold.

Do you get a BSoD or just shut down and restart?

Check your CPU temps and as well make sure your memory is set to XMP in the BIOS. If memory is not set right, it will cause that.

PSU could be an issue but it doesn't seem like it should be as its a decent one.

 

[keithf]

Maybe under load wasn't the correct term. The temp is running swtor on full settings. I know I set the memory to xmp.

There's no bsod its a shut down and restart. When I check the cpu temp in bios (immediately following a crash) it stays around 30c. I'll download coretemp and prime95 tomorrow and get some more accurate numbers.

The psu is around 4+ years old but I'm not really sure how to check if its faulty without getting my hands on another

 

[jimmysmitty]

Getting a known good PSU to test with is the only way. There are voltage meters but they can't tell you how many amps are being put out, which is a very important number when it comes to GPUs under load.

 

[keithf]

Ran prime95 for 10 minutes and my cores were ranging from 55-60 celsius. Gpu is at 100% activity and is sitting around 55 celsius.

 

[daysyang]

I've had similar problem in the past, when playing games my pc would just shut off. It turned out to be psu, under load gpu & psu occt it would shut off but load on cpu prime95 it was fine. Tested my psu and voltages were way below, try testing your psu for failure.