Last weekend my gaming desktop PC started to randomly restart itself. It first did not appear to be after a set period of time or when there was a certain application running, it looked really random. I noted it down and said to myself that I am going to fix the problem in the coming week or next weekend. I was not looking forward to the troubleshooting and diagnosis process.
During the week I was a little busy with other things and I never even got a chance to turn on my gaming desktop PC, so fixing it during the week did not happen. As the weekend came along and I just finished my lunch on a bright and clear Saturday I said to myself that I should probably fix the problem.
Now I have only ever experienced this type of issue before once and have heard a number of different people have the same problem. In my case originally the problem was the PSU and I had to replace it. The issue here was that every single story had a different solution to the problem. Googling the problem will also give you some general and generic answers. The problem could be:
- Hardware related – something could be faulty
- Software related – some virus or application writing to an area of memory it shouldn’t
- Firmware/Driver related – graphical drivers or even deeper problems
I was hoping that the problem was hardware; I had not installed any new software and if I had a virus I would be concerned; Nvidia had put out new graphical drivers but I had not updated any other firmware on my machine. And fortunately enough for me this time around it was only hardware related. My thought process in resolving the problem and narrowing it down to a specific piece of hardware (if it was not hardware related I would have gone through the process of troubleshooting the software, etc.) was to piece by piece remove all hardware component and see if the PC restarts.
I started off by removing and unplugging all my audio equipment and disconnecting my PC completely from the network; not too important and I would not have thought they were the culprit but I need the space and room to work. The first hardware component I removed and test was my GPU. I was hoping that my GPU was not the culprit (it wasn’t) as buying a new GPU right now is out of the question for me. Once my gaming desktop PC restarted with the GPU out I also uninstalled and removed any driver and Nvidia software. Sometimes updates to the graphic drivers causes problems; it has done that to me a number of times. Still the issue persists.
With the GPU removed and all the accompanying software/drivers removed with the issue not resolved, the next logical piece of hardware to remove is the RAM. I have configured 2 8GB RAM sticks, and I first removed the RAM stick that is in Channel A but #2. Issue still unresolved. I replaced the first RAM stick in Channel A #1 with the one I removed first. Low and behold the restarting issue could not be replicated.
To verify that it indeed was the RAM stick, I changed them over to see if it would restart but it didn’t. I was a little confused. After talking to my dad and a couple of my mates on Discord, they said that potentially it was that the RAM stick was just having contact issues and needed to just be reseated. So I played around a little more with my RAM configuration and I could not reproduce the issue. Restarting would normally and reliably occur after at least 10 minutes of my gaming desktop PC being on and logged into Windows 10 when I was testing on Saturday. But now with both RAM sticks (even just having the one identified as the faulty RAM stick alone) installed the issue could not be replicated. This is good news.
I inserted back my GPU, installed the latest Nvidia graphic drivers and made sure everything was up to date, and then just let my PC run a video in the background waiting for it to reboot. It did not happen. Looks like reseating the RAM sticks resolved the problem. Thankfully I do not have to purchase any new hardware and I can still have my PC running with 16GB worth of RAM and not the 8GB if one of the sticks was faulty. So the lesson to be learnt is to methodically go through each of your hardware one by one, noting down any configuration changes and testing to see if the issue can be reproduced consistently. Once the issue has been resolved, see if you can reproduce the issue by reverting the change you had done; once that is confirmed see if your fix when applied again resolves the issue.
Christopher Tom Kochovski 