Computer instantly shuts down when loading games or running FurMark test

[pjnlsn]

Hi

I have a bad problem that my computer instantly shuts down, semi-sporadically, though only when loading games or running FurMark tests. Or more correctly, before FurMark even runs: I'll start it, hit "burn in test," and then instant shutdown, like the plug's been pulled, every single time, though it's sporadic when loading games. And it hasn't done it while a game's running, no matter how long. (least not yet).

I recently updated the video drivers to 314.07

After some research i got the impression that it's probably hardware wise a bad PSU, and i've been thinking about getting a new one, specially if it'll last a while (i've had my eye on the Corsair 600~ watt enthusiast model). But until then, are there any other options? Perhaps even, if it is the PSU being overloaded, a stop-gap measure? I thought it might help if I underclocked the GPU, and i tried lowering the core clock in increments up to -100 MHZ (the default is 740) but no effect on the result of activating FurMark. And i've never seen the GPU temp get above 80, usually 60 when idling.

Any ideas?

Hardware:

Asus m3n78-pro
Athlon X2 4200+
GeForce 9800 GTX+

PSU: Coolmax m-500b
specs on the side of the unit
rails:
150W max on both of these
+3.3v - 28A
+5v - 28A

350W max
+12v1 - 14A
+12v2 - 16A

 

[AMD Radeon]

power outage

coolmax psu .... sorry to say, i am not saying that it is a junk. but it is for very basic pc build. for gaming pc that have GPU. oh boy

corsair CX600 v2
corsair tx 650 v2
xfx 550
xfx 650
seasonic 520
seasonic 620

 

[pjnlsn]

power outage

coolmax psu .... sorry to say, i am not saying that it is a junk. but it is for very basic pc build. for gaming pc that have GPU. oh boy

corsair CX600 v2
corsair tx 650 v2
xfx 550
xfx 650
seasonic 520
seasonic 620

quite

i'm definitely gonna save up for a PSU as my next purchase, probably the corsair tx650, i saw it on newegg for 59.99 after rebate...though i hear the company that does rebates for Corsair is a bit unscrupulous.

in the meantime, i cut all the GPU's clocks by half, and the problem seems to have gone away for now. at least, furmark runs for a good long time now.

 

[benitoll1993]

Hi! after all, what was the cause of this problem? I have exactly the same problem, with an Intel DX79SI, i7 3960x C0 ES, watercooled by AlphaCool Eisberg 240, Quad Channel 16GB DDR3, reference HD7970Ghz Ed. and a Thermaltake SmartSE 730W.

For me, it started happening after a small, supposedly harmless liquid cooling leak, replacement with a new Alphacool Solo block/pump/reservoir, a full disassembly/cleaning/reassembly of the GPU and I had to replace the "thermal sticker" of the heatsink above the CPU because it was loose, so I don't have a "I changed this specific thing and the problem started happening", but several possible causes of the problem and no way to test the components by myself.

I think it must either be the PSU or some faulty VRM/capacitor somewhere, I think it can't be thermal related because it doesn't happen with continuous load but just when it switches from no/low load to mid/high load... if it doesn't crash instantaneously, it's very rare that it crashes while continuous high load (but also happens, rarely).