1

Edit: This happens consistently in games but also occasionally (once every 4 or 5 days) when having trivial apps like Skype open.

So I built a slightly new PC back in May. The upgrade consisted of a new CPU (i5-2500k), a new miniatx case, a motherboard (p8p67-m), some ram and a PSU (an OCZ 600W). The other stuff such as the 5850 I put into this one as I had absolutely no problems with it along with the hard drives.

However, as soon as I started playing a game from 10 minutes to 2 hours, I would get an abrupt restart. No minidump created/found as far as I can see, automatic restart disabled...just a complete hard reboot. This happens within the 10min - 2hr range in a game but without a game, it does happen to only maybe once ever 4-5 days. For example, I was just on skype the other day and for no reason it did the same hard reboot.

I have done memtest86+ up to around 93% (?) with no signs of error. I then messed around with drivers and did a new install of windows 7 with no help. I've tried different games with the same behavior. I switched out the PSU finally with a new Corsair 700W and still, the same issue. I also changed the case to a different thermaltake mini-atx one.

Thinking that maybe the microatx case (overheating) was a problem, I logged the temps before a complete reboot and here they are, I don't think they look that crazy:

CPU

RealTemp:

  • Date : 9/11/11
  • Time: 23:42:00
  • C0(F): 138
  • C1(F): 136
  • C2(F): 133
  • C3(F): 136

GPU

  • Date: 2011-09-11 23:42:23
  • GPU Core Clock [MHz]: 725.0
  • GPU Memory Clock [MHz]: 1000.0
  • GPU Temperature [°C]: 79.0
  • Fan Speed (%) [%]: 33
  • Fan Speed (RPM) [RPM]: 2324
  • GPU Load [%]: 87
  • GPU Temp. #1 [°C]: 80.0
  • GPU Temp. #2 [°C]: 83.0
  • GPU Temp. #3 [°C]: 82.5
  • Memory Usage (Dedicated) [MB]: 276
  • Memory Usage (Dynamic) [MB]:40
  • VDDC [V]:1.0875
  • VDDC Current [A]:34.8
  • VDDC Phase #1 Temperature [°C]: 68
  • VDDC Phase #2 Temperature [°C]: 68
  • VDDC Phase #3 Temperature [°C]: 67

So to sum it up, hard reboots, no minidump that I can see, kernel-power error only in event log, the older 5850 worked fine in my old build and this happened very quickly once I switched over to the new hardware. Meaning, it wasn't like it was a gradual thing which occurred after two weeks stable but rather right away. I only bring it up now since I've gotten to the point of frustration with this.

5
  • This is most probably a hardware failure probably temperature-related, since it happens more on games. This is hard to diagnose specially when you don't know when it would happen. I would try replacing hardware components, starting with the power supply. On most similar scenarios I've seen, it's usually either a mobo or a power supply hardware problem (poor components).
    – m0skit0
    Sep 12, 2011 at 13:47
  • Yea but I already replaced the PSU once and this time with a corsair I mean, it would really be amazing if both were shoddy.
    – Ilya
    Sep 12, 2011 at 13:51
  • Check the bios for a temp log.
    – Moab
    Sep 12, 2011 at 14:10
  • 83° on the GPU doesn't sound nice. (180°F). Coupled to the fact that it only happens when you play games...
    – b0fh
    Sep 12, 2011 at 16:11
  • This may sound completely dumb but when GPU overheats, shouldn't I see some weird artifacting occurring? Not a complete restart?
    – Ilya
    Sep 12, 2011 at 16:30

1 Answer 1

1

Solved: One of the two sticks of ram was causing the issue.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .