PC keeps restarting without warning.

[TickleMyElmo]

So my pc has shut down and restarted twice in the last week. I have no clue what it is, PSU maybe? What do you guys think?

 

[BuRgLaR]

So my pc has shut down and restarted twice in the last week. I have no clue what it is, PSU maybe? What do you guys think?

What are you doing when it restarts? Gaming, Browsing internet , Video editing ?

 

[jmh547]

My first guess would be psu but it could be tempurature also.

What are you doing when it restarts?
What do your temps look like?
Do you get any error messages?

 

[TickleMyElmo]

It's not temps. Realtemp never goes above 60c. I was gaming both times.

- - - Updated - - -

Gonna turn up the cpu voltage and see if that helps.

 

[wagex]

if it just powered off its not a stability issue. id check your motherboard and psu... also your case button.

just incase, make sure the powerstrip isnt under your feet and you arent bumping the reset button.. lol.

 

[TickleMyElmo]

Are there programs to check the health of the PSU and/or MB? I have reason to believe it's the PSU, but not sure.

 

[trents]

My first thought is that it is the PSU. The only good way to check PSU health is with a multimeter when the PSU is under load. There is a description of how to do this in a sticky located in the PSU section.

 

[EarthDog]

Yep, load up Unigine Heaven and run a CPU stress test of your choice at the same time and see if it still happens (you can check the PSU voltages while doing this). If it does, then its likely a power issue of some sort, be it the PSU/loose connection, something along those lines.

 

[TickleMyElmo]

Can I use Valley?

 

[EarthDog]

Of course. Just something gpu heavy.

 

[TickleMyElmo]

Ok, ran Valley on highest settings and cpu at 4.2. After about 4-5 minutes the cpu reaches 75-80c, GPU stays under 65c. There were no artifacts or anything. The CPU was getting over 75c which I set on realtemp for the alarm to go off. But still no crash.

 

[knoober]

I had a similar problem. Restart at random intervals (although more often when stressing) and for no reason. I dont know the specifics behind it but trading gpu fixed the issue. Test with a spare psu maybe?

 

[EarthDog]

Ok, ran Valley on highest settings and cpu at 4.2. After about 4-5 minutes the cpu reaches 75-80c, GPU stays under 65c. There were no artifacts or anything. The CPU was getting over 75c which I set on realtemp for the alarm to go off. But still no crash.

to be clear, you ran valley and a cpu stress test (which one?) At the same time? How long did you let them run together? For close to how long it takes to reboot?

You lowered your overclock as well?

Does it happen in terms same game?

 

[trents]

I would next look at RAM issues. Take out all but one stick of RAM and test each one in turn in each slot. I would also probably start by running memtest86+ overnight. One advantage of that is that memtest runs off the CD in a DOS environment so you would be bypassing any protected mode Windows drivers.

 

[TickleMyElmo]

Earthdog, I ran p95 small fft's and Unigene Valley (highest settings) and about the 4-5 minute mark the cpu was getting over 75c. There however was no lag or crash. I have not lowered my cpu overclock yet. It's at 4.4 at 1.35 volts.

 

[EarthDog]

Well, seemingly not power. Try mem test now like trents said.

 

[bob4933]

Pull your event viewer logs. There should be something there if it's memory or gpu related.

 

[Mr.Scott]

This ^^^^^
It's very rare to have a memory or GPU issue that will just restart a machine without BSOD or a log entry. Generally, no BSOD = power or heat issue.

 

[TickleMyElmo]

I checked event viewer, says kernal power.

 

[Mr.Scott]

So , kernel power ID 41. That means windows did not shut down clean. Most often because of power outage.
So you're right back to the same thing, heat or power.