I have a Windows 8.1 custom PC that I built about 6 months ago. It worked great for several months with no problems, but about three months ago it started to reboot randomly.

My computer works fine for about twenty or thirty minutes, but then the screen will go black and the computer will restart. After the first reboot, my computer will restart before reaching the Windows loading screen and will keep restarting until I turn it off. If I turn it off and wait twenty minutes before turning it back on, my computer will run normally but after twenty minutes it restarts and everything starts all over again.

I noticed the computer only reboots when its running in Windows. I ran the computer in BIOS for several hours and it didn't reboot. I also noticed that sometimes it won't reboot if I just let my computer sit idle in Windows, but it always reboots when I run a program or app, even if the app isn't resource heavy.


Motherboard: Asus M5A99X EVO R2.0
CPU: AMD FX-8320 Eight-Core
Memory: Corsair XMS 3 DDR3 RAM (2 Sticks, 4GB each)
Graphics Card: AMD Radeon R9 270
Hard Drive: Hitachi HDS721050CLA362 500GB
Power Supply: Corsair Enthusiast Series TX650M (650W)
Optical Drive: DVD/CD-ROM Combo Drive
Case: Corsair 500R

I've done a few things to figure out what's wrong. I ran Memtest86+ to check my RAM, and it passed with no errors. I also monitored my CPU's usage, load and temperature using Task Manager and CoreTemp, and everything was normal right before the computer restarted. My CPU temperature was normal, always near 30°C to 39/40°C (the maximum temperature for my processor is 90°C) and the load was at most about 30%, so I don't think my CPU is the problem.

I ordered a psu tester to check my power supply. If the PSU is fine, then I might send my motherboard in for an RMA. I also thought about testing my hard drive, or running Prime95 but I haven't yet since my computer won't stay on for more than 30 minutes.

Does anyone have an idea what the problem is? Can I do anything else to test my computer?

  • read this KB article: support.microsoft.com/kb/2028504 – magicandre1981 Jun 1 '14 at 5:43
  • @magicandre1981 Thanks for the help. My computer's event log does have several event ID 41 errors from all the times it has restarted, but all the values (like bugCheckCode) are 0. From what I read, this confirms that the reboot is caused by hardware failure right? – kehmoto1 Jun 1 '14 at 5:54
  • Do you get the same issue if you start Windows in Safe Mode or perform a clean boot? – and31415 Jun 1 '14 at 12:44
  • try all steps from scenario 3 that are written in the KB article – magicandre1981 Jun 1 '14 at 17:23
  • I've had a screwy BIOS cause this. The motherboard was old enough to have a revised BIOS I could use. Also had hardware compatibility issues that caused this. Point is, every thing may be working exactly as it was designed, but things still could be a problem. You might try loading a different OS, or swapping hardware and testing between each swap to isolate the cause. – Damon Jun 2 '14 at 6:33

90 deg is way high and above the safe operating temperature for this chip. Make sure you've got a good fan and that you have good contact with thermal paste and absolutely no air pockets in it.

Check the Event viewer, look for errors or warnings surrounding the kernel power loss events (they'll be there, because of unexpected reboot). There may be another issue, such as a bad driver or Windows Update, it may be that you need to update your bios. I'm fairly confident your CPU is spiking in temp and forcing a shutdown.

The boot loop you experience may be the bios temp monitor forcing a reboot, but this is strange because the normal behavior is to force a full power down, not a system restart. But, who knows, maybe this is the programmed action. The boot loop may also be caused by a Windows update (I can't recall which) caused precisely the same problem for me (and many others) on Windows 10 Home on various hardware, and if that is the cause, you can stop at least half of this issue by disabling the the Fastboot option in your bios. See here for how to do that on your system.

While you're poking around in the bios, find the area that has temp monitor nfo and check your limits, then ensure that they're in line with AMD's recommendation for your specific chip.

  • ugh, I didn't realize this question is so old. This site is different from other SE sites, it necros threads constantly. – user72945 Feb 19 '16 at 18:15
  • Superuser works exactly how every other SE site works, it uses the exact same system and source code. Your user preferences are just different, so sort the questions, only reason this was on the front page was because of your settings. It was then placed on the front page because of this answer. – Ramhound Feb 19 '16 at 18:48
  • @Ramhound Well it's a good thing I guess, because apparently nobody of 2854 visitors over 2 years realized that a 90 degree CPU temp is bad. – user72945 Feb 20 '16 at 0:00

Have you tried disabling the "Hybrid Shutdown" feature as described in this article?


I'm having the exact same symptoms myself, and I tried the fix noted in the article and it seems to have helped. However, it's only been a few hours so I'm not confident this is the fix I've been looking for.

  • 1
    Welcome to the site. This looks like it might be a good suggestion. External links can disappear and if that happens, your answer wouldn't be useful. In addition to the link, summarize the key points of the article or even paste sections of it into the answer. The question has been around for 5 months, had 1,200 views, but no other answers. If you make this change to yours, it might get some upvotes. – fixer1234 Nov 3 '14 at 4:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.