Computer powering off whilst gaming

ttreaders

ttreaders

ttreaders

Associate
Joined
28 May 2007
Posts
656
Hey guys,

I have a system bundle bought from ocuk back in Jan 2012

Krypton Z68 670i Intel 2500K 3.30GHz @ 4.60GHz
MSI Z68A-GD55-G3 Intel Z68 (Socket 1155) DDR3 Motherboard
Kingston HyperX Genesis Grey 8GB (2x4GB) DDR3 PC3-12800C9 1600MHz
Geforce 660 graphics card
400w Corsair PSU - (I know this seems on the low side but my new CPU/GFX actually used less power than the previous so I stuck with it)

Initially when I bought the bundle I was getting BSOD's and it was due to the CPU not having enough power. Under advice from ocuk staff I increased the voltage and it became stable. I've been playing for the last year or so with no crashes.

I have just started playing the Wildstar beta however and after playing for a few hours my computer instantly shut down and then rebooted.. no error messages or system log as no actual error occurred as such, it just powered down. After rebooting I logged back into Wildstar and within a minute it did it again.

I initially wondered if it was overheating, checked temps and they are fine (cpu is normally 45-55 whilst playing) and gpu is well under safe working temps.

I increased the CPU voltage again slightly to see if this game was just more demanding than others I had played and maybe that's why it was powering down? (a stab in the dark on my part). I increased it to 1.4v which has led to the temp of the CPU raising to the late 60's whilst gaming though I managed to play for 7 hours with no more powering down.. could be a coincidence however.

So my question is, am I right in thinking that the PSU could be failing? Or could it be that the CPU just wasn't getting enough juice for the demanding game?

Thanks!
 
Try the other way, drop the overclock down slightly, try 4.4ghz with less volts.

Wazza normally has some good advice for sorting the voltages out.
 
It sounds like one of the PSU's protection features is kicking in under load and it might just be down to age, or that the newer components require higher currents (though less total power perhaps) than the previous ones and this is stressing the PSU.

My old, but very reliable, Enermax 620W couldn't deliver enough amps to let me overclock my GTX680 (if it went above ~105% power then I'd get a driver crash). My new PSU (Superflower 1000W) happily runs the GTX680 at much higher speeds with no crashing (Afterburner shows peaking up to 137% power now).
 
Try the other way, drop the overclock down slightly, try 4.4ghz with less volts.

Wazza normally has some good advice for sorting the voltages out.
Click to expand...

Thanks was thinking about that but problem is I don't know how.

It sounds like one of the PSU's protection features is kicking in under load and it might just be down to age, or that the newer components require higher currents (though less total power perhaps) than the previous ones and this is stressing the PSU.
Click to expand...

Thanks for the reply, do you think the PSU could give up soon then?
 
ttreaders said:
Thanks for the reply, do you think the PSU could give up soon then?
Click to expand...

Not necessarily, and especially so if it's a quality unit. The protection features should just kick in more often. It could be over-temperature protection (in which case you're also likely to notice the fan spinning faster and/or more often), or over-current protection (not sure what the outward signs of this would be beyond that which you're already experiencing).

It is generally the capacitors aging (or the opto-coupler) in PSUs that causes failures/unreliability or reductions in performance after a long service life, and the upshot of this would be that the PSU is having to work harder than it used to in order to deliver a certain amount of power. This will create more heat therefore exacerbating the problem.

A quality PSU shouldn't fail without any prior signs.

If you have access to another PSU that is known to be good I'd recommend swapping it out for some testing.
 
So just to clarify Havana, are you saying that the PSU could just be getting old and 'worn' so it's struggling to provide enough power when running demanding applications hence the crashes? But on the other hand if me increasing the CPU voltage stops the issue then it rules out the PSU from being faulty?
 
ttreaders said:
So just to clarify Havana, are you saying that the PSU could just be getting old and 'worn' so it's struggling to provide enough power when running demanding applications hence the crashes?
Click to expand...

Yes, this is a possibility. Generally speaking your computer's GPU is the most power hungry component of the system drawing more amps than any other single component. It could be that this game is more demanding and therefore your GPU is drawing more current than it has before.

If that current exceeds that which the PSU is capable of supplying (either because it's old, or just could never supply this much current in the first place) then one possibility is that your system will crash.

ttreaders said:
But on the other hand if me increasing the CPU voltage stops the issue then it rules out the PSU from being faulty?
Click to expand...

It may be, however, that this particular game is simply more demanding on the CPU than prior games and therefore creating instability, causing the system to crash. In this case, increasing the Vcore could help to solve the problem.

Windows should create some event logs that may help you trace the cause of your problem. You need to find the "error" events immediately prior to the system shutting down.


Increasing the Vcore and the problem going away might be unrelated. If you reduce your Vcore does the problem occur again? Alternatively, if you cancel the overclock completely and run the system at stock, do you still get issues?
 
You must log in or register to reply here.
Back
Top Bottom