[SOLVED] PC crashes 10 times a day due to unknown reasons

Feb 9, 2021
2
0
10
Hi,

What I've done:
My PC crashes 10 times a day and I haven't been able to figure out why. I'm not a tech-savvy guy but I still think I have done my part in trying to figure out what the issue(s) might be.
I got this PC in 2012 and it has been running smoothly until December of 2020 when it suddenly starting crashing at random moments.
The last two months the crashes became more and more regular to the point where I practically cant use it anymore.
I installed Speccy some time ago to try and find out what the problem could be. Everytime, my computer crashes my CPU is below 25% usage, my memory is used about 25% and the GPU is used about 10%.
Additionally, I installed Core Temp to figure out if overheating was the issue. My fan at the time made A LOT of noise so I bought a new water cooler. Though the noise has stopped, the crashes haven't.
The temperature always maxxes out at 55 degrees celsius, but normally lies in between the 25 to 45 mark, which to me doesn't sound like an issue.
I then began suspecting that my, at that point, fairly old power supply could be the culprit. I opened my PC and cleaned it from top to toe making sure there was no dust or dirt to irritate it.
This includes the entire computer from CPU, RAM, Power supply, fans etc.
Still didn't work. I then completely rebooted my PC from scratch to make sure it wasn't some stupid Windows malfunction causing the error, which it turned out it wasn't.
I then upgraded my BIOS from a 2015 version to a 2018 version. Still no luck.

At this point, I don't know what to do anymore. My CPU seems to work fine, same goes for my RAM and GPU. The computer clearly isn't overheating and Windows works as intended. I'm lost.

How the crashes happen:
When the crashes first started happening, it was usually when I had a fairly big amount of stuff open on the PC.
It might've been a GPU intensive game along with maybe 2-3 tabs on Google Chrome and then Spotify and Discord running in the background.
But as time progressed, the crashes worsened. If a big explosion happens in a game, it instantly shuts down the PC. If I have too many Chrome tabs open it shuts itself down after 10-20 minutes.
And the most common of them all is when I have, let's say a Google Chrome tab, open on one screen and then the millisecond I click on my second screen it crashes.
I can move my cursor around on the page without problems but the very millisecond I press Mouse1 it shuts down.

To me this sounds like two issues. The "big explosion in a game part" would suggest my GPU cant follow the demands of the game and the crashes in Google Chrome suggest that it's a RAM/memory issue.

I am completely lost to what to do next.
I would highly appreciate if any of you guys posted some possible solutions.
I will post my specs below;

Thanks in advance.


PC Specs:

CPU: Intel Core i7 4790K
RAM: 16,0GB Dual-Channel DDR3
Total Memory Slots: 4
Used Memory Slots: 2
Free Memory Slots: 2

GPU: NVIDIA GeForce GTX 970
Motherboard: ASUSTeK COMPUTER INC. MAXIMUS VII RANGER (SOCKET 1150)
Hard drive: Samsung SSD 850 EVO 500GB
 
Last edited:
Solution
Check in event viewer for any clues or critical errors.
Did the system create a dump file at BSOD?. If so then the dump file needs to be analyzed.

You need to check your System config. If it runs in safe mode, it's most likely a driver or service issue but not 100%. Check services and try different configs.

Configuration problems occur with some software packages when first installed. The user must enter certain parameters into the program to match its capabilities to the system's actual configuration. ... If these parameters are set incorrectly, the software cannot direct the system's hardware properly and an error occurs


Now check the PSU.
You can check your rail voltages using HWInfo64 when the system is under load...
OK lets rule out thermal or hardware issues for now and check if it's driver related.

Is the GPU driver up to date?
Have you tried booting into safe mode?
Check in Device Manager for any yellow triangles indicating a bad driver.
run SFC /SCANNOW in an elevated command prompt to check for any system32 corruption.
List your PSU?

Many things can be the cause so be patient and report back on the above.
 
Feb 9, 2021
2
0
10
OK lets rule out thermal or hardware issues for now and check if it's driver related.

Is the GPU driver up to date?
Have you tried booting into safe mode?
Check in Device Manager for any yellow triangles indicating a bad driver.
run SFC /SCANNOW in an elevated command prompt to check for any system32 corruption.
List your PSU?

Many things can be the cause so be patient and report back on the above.
1) Is the GPU driver up to date?
- Yes, the driver is up to date.

2) Have you tried booting into safe mode?
- I just tried it and I did not see any problems whatsoever. I tried opening up as many things as I could to test the RAM and it worked fine. What does that mean?

3) Check in Device Manager for any yellow triangles indicating a bad driver.
- No yellow triangles

4) run SFC /SCANNOW in an elevated command prompt to check for any system32 corruption.
- I ran it and it did not find any problems

5) List your PSU?
- 550W , 80 PLUS Bronze
 
Check in event viewer for any clues or critical errors.
Did the system create a dump file at BSOD?. If so then the dump file needs to be analyzed.

You need to check your System config. If it runs in safe mode, it's most likely a driver or service issue but not 100%. Check services and try different configs.

Configuration problems occur with some software packages when first installed. The user must enter certain parameters into the program to match its capabilities to the system's actual configuration. ... If these parameters are set incorrectly, the software cannot direct the system's hardware properly and an error occurs


Now check the PSU.
You can check your rail voltages using HWInfo64 when the system is under load.
Try to conduct a stress test using AIDA64.
You have a bronze rated PSU (basic and wattage is minimal for your setup) so should any rail voltages be out of spec ( +or- 5%) then swap out the PSU with a known working unit of the same or higher Wattage.
 
Solution