Question How to tell if my CPU or GPU has gone bad?

Oct 18, 2019
Good day all,
during a gaming session (nothing too hardware demanding...WoW), my PC crashed and left me with a red'ish striped screen (after further crashes also in blue or black). Powering on again worked after a couple of tries, but now my PC won't post at all, can't even get into bios. This has never happend before, also I did not change anything software/hardware related for a long time.

Naturally I thought it might be my GPU that got fried (it's a custom built PC and ran smoothly for around 6 years), but then I tried a lot of things and I'm getting more and more confused.
What have I tried so far:
  1. CMOS reset (cut power completely, removed GPU and RAM and removed the CMOS battery for 5 minutes while keeping the power button pressed for half a minute or so) --> no changes, still back screen (btw I got 2 screens, no post on both of them)
  2. I've tried removing hardware components bit by bit to see if there's another faulty part, but no changes
  3. my CPU has no iGPU, so plugging my VGA cable directly onto my mainboard won't help me rule out any GPU issues
  4. I removed all the dust from the fans (wasn't too bad considering it hasn't seen any light for 6 years)
  5. every 20 tries or so my PC boots normally (!) - no noticable performance issues...but when I launch a game, it crashes my system again after playing for a couple of minutes (again, without any perofmance issues)
  6. I started a GPU benchmarking test with FurMark and let it run for 2 minutes on highest load, and my system did NOT crash...
  7. I started some loggings, hit up a game again and it crashed again after ~2-4 minutes and now I'm back at my system won't POST/do anything. Fans and everything run smoothly, but nothing else happens
  8. I changed RAM with new ones, no changes
  9. I changed PSU with a new one, no changes
My hardware components: HWInfo

What logfile can I provide?
  • HWInfo sensor status over 20 minuts while idling part 1 part 2 part 3 part 4 part 5
  • GPU_Z log before and during the crash while playing WoW (when the GPU clock switches to 1000, that's the moment the game launched): part1 part 2
  • GPU_Z log during the FurMark GPU benchmark test: part 1 part 2
Unfortunately I'm terrible at interpreting such logs...can someone find anything suspicious in there that points me in the right direction? Now that I can exclude a RAM and PSU related issue, it's down to the more costly issues (GPU, CPU or MoBo?)



With all things posted in your thread, you forgot to mention the specs to your system. List them like so:

Also, you should mention what the PSU's used were, both old and new.

To me, it looks like your GPU isn't receiving power on demand(faulty PSU) or that your GPU is on it's way out. To also note, the GPU you've shown in your screenshot, apparently has been dying for a lot of people, from either thermal issues, people with bad quality PSU's or simply using the wrong drivers/overclocks.
Oct 18, 2019
I posted my specs as a screenshot from HWInfo...but here also in text form:

CPU: Intel Xeon E3 3.40 GHz
Motherboard: ASRock H97 Pro4
Ram: 8 GB DDR3
SSD/HDD: multiple SSDs, one HDD
GPU: AMD Radeon R9 280X
PSU: Antec Ne550 W
OS: Win 10

Also I changed my PSU already (was a be quite! 450 W PSU, now upgraded with a 550W, but still the same issue). Also I got all of the latest drivers, even upgraded the BIOS when I was last able to start my PC.
Last edited: