PC random reboot

[Toomt]

Hello.

First of all let me start by saying that this problems only seems to happen in Crysis 2 demo now. I started having random reboots a few weeks back when playing Mafia 2 but i fixed it by underclocking my GPU because it seems it was overheating. (Not really sure if the heating was the problem or if it the GPU was sucking up too much juice along with the CPU).
Anyway, this time with Crysis 2 (demo) the reboots are totally random. Sometimes it happens a minute after the action starts, sometimes 5-10 minutes and sometimes i can play 3-4 full matches and quit without a crash. (It's a multiplayer demo)

My PC specs:

Motherboard: Asus P5Q Pro
CPU: P Q9550 (not overclocked)
RAM: 2x 2GB DDR2 800MHz - Working in Dual Channel
GPU: XFX GTX260 Black Edition (factory overclocked)
PSU: 580W Eurotech Silent Pro (PSPA580PSA)

Software:

Windows 7 Professional SP1 - Fresh install
GPU drivers up to date from Nvidia website
DX11

Heat: CPU goes as high as 78ºC in one of the cores, and 64ºC overall. The GPU goes as high as 70ºC.
Factory default GPU settings are Core Clock at 680MHz which i lower to 650. Shader clock at 1466 which i lower to 1400 and Memory Clock at 1000 which i lower to 950. These values mean that the temperature is stable at 70º with the fan at 85%. (I use that value while playing)

Ran memory tests, no problems found. Disabled the auto reboot feature in Windows System and i still don't get a BSOD.
I don't have a firewall and removed the antivirus temporarily but the problem persists.

I was running AIDA64 configured to save a log file in htm which you can find in this URL: http://www.mediafire.com/?05h7he4dtco79ae
I don't see anything there that can give me any information about the crash, but then again i'm not a computer engineer.
The last line on the log could be anywhere from few miliseconds from crash up to 5 seconds.

Meanwhile i have installed a new CPU cooler because the temperatures were too high. I managed to decrease the CPU overal temperature to 45º as opposed to the 64º i had before (under stress) but the problem persists. The updated AIDA64 log just before the last reboot is here http://www.mediafire.com/?1hdyn3osevv89bm
Also i ran all the OCCT tests and i get a reboot few seconds after the PSU test starts, so if i had to take a guess i would say this is where the problem is.

I would also like to ask in advance, if it is indeed a PSU problem is there a way of optimizing it in the BIOS (changing voltages or something) to prevent this from happening or is getting a new one the only option?


Thank you for your time. Any help will be greatly appreciated.

Regards,

John

Edit: In OCCT, the crash occurs after the PSU test starts. The others run fine.

 

[JKatwyopc]

I ran your specs through a couple of PSU calculators (Corsair and ASUS) and they indicate that you need about 550 watts. Thats assuming you have one hard drive, one DVD/CD drive, a couple of USB devices. If you have more than one hard drive or multiple GPU's, lots of USB devices (5 or more) you might need a bigger PSU.

More info about your hard drives and other devices would be helpful in determining whether or not your PSU is sufficient.

BTW even though Win7 has DX11 in software, your GTX260 based card does not support DX11. In order for DX11 to work it needs to be supported by both software and hardware.

 

[frozenlead]

You're probably going to need to replace the PSU. Even though the unit you have is 580W, I doubt it could push that much. Buy a higher-quality brand, like Seasonic, Corsair, or Antec (these aren't the only options).

 

[Toomt]

Following the "lack of juice" theory i disconnected all non essential components like the 2nd hard drive, the external hard drive and the DVD. Furthermore i underclocked my GPU for about about 100MHz of each value.
I didn't run the OCCT PSU test but the reboots while playing Crysis 2 seem to have stopped.

Edit: Tried the same but without underclocking and it rebooted after a while. Is the Corsair TX-650W a good choice for my system for 89€ ?