EDIT: Solved by new PSU.

I have an old PC, now used basically as an office machine:

  • CPU Intel dual-core E8200
  • Mainboard MSI P35-Neo2
  • GPU GeForce 8800GT
  • HDD Seagate Barracuda 2TB
  • RAM 4x 1GB DDR2
  • PSU Fortron 500W
  • OS Windows 10

Roughly a week ago it started to restart when playing any video or audio. At first, it played for about 10-15 min, after the restart, it lasted for just about 1-3 min. Since then, the issue gets systematically worse. Few days ago, it started to happen while browsing (no video or audio playing), now it randomly restarts even when completely idle (no application launched) and then it doesn't even completely boot and gets stuck in boot-loop. Only when I turn it off for a moment, it is able to boot completely, but it is all the same after few minutes. Also, it always restarts if I put it in sleep mode, that's the only way I can cause it on purpose.

There is no BSOD, no problem in the Event Viewer, just the inevitable event ID 41.

Two things happened at similar time as the first restart - it got infected by a bitcoin miner (already gone I believe) and I installed Realtek audio drivers (already uninstalled). I'm not really sure about the exact sequence of these actions.

I'm pretty sure it's not an overheating issue. The PC is on the ground in a cold room, I use Noctua fans for both the case and the CPU and I regularly clean the dust. I use Open Hardware Monitor all the time, the temperatures are fine.

I tried the following:

  • ran antivirus, malwarebytes and others, no problems
  • ran memtest86
  • ran GPU stress test Furmark
  • disabled auto restart
  • disabled sleep mode
  • switched SATA cables
  • switched HDD
  • removed all the RAM sticks one by one
  • reseated GPU
  • disabled audio in BIOS
  • returned overclocked settings to default
  • visually inspected the mainboard capacitors

The one and only thing which seemed to work was switching the power outlet/power cable. That was yesterday and when I did that, I was able to play video for about an hour without any issues, so I thought it's solved. Today, even in the "correct" outlet with a new power cable, the restarts are as bad as they ever were.

My question is: is there anything else I can try before I spend money on new stuff (new PSU is where I plan to start)?

  • 2
    I have used OCCT to test my computer when it had a random reboot problem. The data/graphs it saves helped me determine there was a problem in one core of my CPU. However, my issue was a reboot every day or two, I'm not sure this would help with as often as yours does. Your plan to buy a new PSU is what I would do at this point, as that sounds the most likely culprit. You might try plugging it in somewhere else in your house just to make sure it's not a home-wiring issue. – techturtle Nov 21 '17 at 13:52
  • 1
    you can start Windows in safe mode to see if the problem persists. Just to rule out the operating system's fault. Or boot a Live Linux from USB and see if the problem persists. – Joe6pack Nov 21 '17 at 14:29
  • So I installed OCCT, ran the CPU test for 20 mins, no problem. When I turned on the PSU test, PC restarted practically immediatelly. I will try it once more (i has to be off for a while to even start) just to be sure. – pn21542 Nov 21 '17 at 14:56
  • 2nd try, OCCT PSU test ran fine for 15 mins. Only after I stopped it the restart happened again. When I think about it, I don't remember it happening under sustained high load. It typically restarts under low/medium load or some load variation (i.e. going into sleep). – pn21542 Nov 21 '17 at 15:40
  • @Joe6pack Ran Windows in safe mode, exactly the same behaviour. – pn21542 Nov 22 '17 at 2:13

Seeing that you have run a couple of test there, but still no cigar i would sugest the next test set:

Go barebones. Remove everything unnecessary, so the power consumption is low. If no reebot on idle, then that's it. Then test it with everything in on max load, it should fail shortly.

In the past i thought that i've fried my MB, had random reboots all the time. Turns out it was just a faulty PSU(twice, cheap psu :D). I did notice that everytime i pulled something out like a HDD, or a GPU, the session would last longer as long as i didn't use lots of resources. Then one day it died, so my this friend lend his psu and all was working like the day i bought the pc. Got me a brand new Cooler Master PSU and no problems till now(it's a 950 i7, which it has been powered on most of it's life)

  • It is barebones already, optical drive and 2nd HDD disconnected. That's it. Audio and network are on-board only (and audio disabled). No effect unfortunately. – pn21542 Nov 21 '17 at 15:33
  • @pn21542 Do you have any tool to get a read for volts or amps from your psu? If so, you should be checking volts accoring to THIS, if the messures are not those, then it's the PSU. Ask a friend to lend you their PSU and test your hardware. – dmb Nov 21 '17 at 19:03
  • 1
    A PSU failure can happen to a specific rail/voltage, so it can fail in such a way that peripherals don't play any part in the failure. That's not usually the case, though. The video card take a lot of power, so that could have failed. If you have a card that doesn't take the specific video card plug from the PSU, you might have a stable computer. – computercarguy Nov 21 '17 at 19:06
  • 1
    It looks like your video card requires a minimum 425 watt PSU, so with your other power requirements and the (possible) age of the PSU, it's looking more and more like the PSU is the culprit. BTW, if it's the Fortron made by FSP, I've not had good luck with the FSP brand. – computercarguy Nov 21 '17 at 19:07
  • @dmb Let's say I don't like to get my hands dirty, so no, no tool for measuring volts and amps :-) – pn21542 Nov 21 '17 at 21:10

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.