0

Recently I noticed my computer blue screening more than usual. Upon inspection I found my GPU was hitting 83C, pretty toasty. I decided to replace the stock thermal paste with Arctic Silver MX-4 after that my computer had increasingly more blue screens (GPU still hit max of 83C). I've come to the conclusion I've boned the GPU. It seems to only crash when under some sort of load, I've disabled hardware acceleration in chrome and it seems to be alright for now. I've tried downgrading and updating my drivers, reinstalled Windows twice. I have tested my memory with Memtest86, 0 errors with rated XMP. I've re-applied the thermal paste for a hail marry. Here is a picture of WhoCrashed's mini dump analysis. If anyone could give any suggestions, that would be great.

Specs:

OS: Windows 10
CPU: Ryzen 7 2700X - 8C/16T
RAM: G.SKILL Ripjaws V Series (2x16GB@3200MHz)
Mother board: MSI X470 Gaming Pro Carbon
Video Card: MSI 8G 1080 Armor OC
Power Supply: EVGA 650W - EQ (80+ Gold)
1st Hard drive: Samsung Evo 970 - 1TB NVMe M.2 (OS Drive)
2nd Hard drive: Western Digital Red - 2TB 7200RPM
3rd Hard drive: Western Digital Blue - 1TB 7200RPM
Internet Adapter: Onboard Ethernet
Case: be quiet! Silent Base 601
Cooler: EVGA CLC 360mm - 3x Sycthe Slipstream 120mm
  • What do you want us to say? Try a different graphics card, if the fault persists then something else is at fault. – Michael Frank Nov 8 at 3:44
  • I don't have a second nvidia card – diamondpumpkin Nov 8 at 3:55
  • Your BSODs are completely random. The most likely thing is faulty memory. Testing is not enough. Run on one module or the other until you know for sure it’s not memory. Make sure the memory is actually running at the proper voltage and timings according to the manufacturer and make sure all overclocking related settings are turned off. Modern motherboards have a ton of silly overclock settings that are typically on by default. Turn all that garbage off. – Appleoddity Nov 8 at 4:21
  • @Appleoddity None of it is overclocked by default, aside from XMP everything is stock. The GPU managed a 10 minute furmark test just fine. I'll fiddle with the memory sticks some more. – diamondpumpkin Nov 8 at 4:50
0

Now, I tried a boat load of stuff to get things working again. I managed to do it, some how. I believe the main cause of all my crashes was XMP, it was working for a while but it suddenly decided to quit. The memory modules seem to be working, I'm going to see if G.Skill will replace them anyway.

  • Why not test your memory to see if an error is detected. 83 on your GPU isn’t that high it certainly isn’t high enough for hard crashes – Ramhound Nov 8 at 17:09
  • @Ramhound I did, only gave errors with XMP. – diamondpumpkin Nov 9 at 13:31

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.