Extremely Weird and Annoying Crashes; Most Likely GPU Related

[Damir Halilovic]

Heya,

For a while now I keep getting extremely weird crashes. There is no blue screen, the PC just restarts. The symptoms are as follows:

• 
  1. It can happen during gaming, or, as was the case earlier tonight, right after rebooting from crashing.
  2. It's not a heating issue (neither the CPU nor GPU go over 60 degrees celsius)
  3. It happened on Win 8.1 and is now happening on Win 10. I solved it a few months ago with what I thought was a BIOS update, but it returned after I did a fresh windows install (this is why I believe it's software related)
  4. I did 6 passes with memtest86+ booted from USB, it came back clean. I did furmark for 4 hours with no issues. I scanned for malware with both the windows tool and bitdefender, at startup and in safe mode, all is clean.
  5. https://www.dropbox.com/s/lr1e2xjqqbpaq99/061415-8531-01.dmp?dl=0 <----Memory dump
  6. Windows as well as all drivers, are up-to-date.
  7. Most of the time event log shows an error 41, unexpected power loss. A few times it gave an VIDEO_MEMORY_MANAGEMENT_INTERNAL error. Microsoft's BugCheckCode reference site lists possible error parameters, but this one (2e) is nowhere to be found for this particular error.

Any help would be greatly appreciated!

Best regards,
Damir H.

 

[SinxarKnights]

What are your system specs?

A sudden restart usually indicates a PSU issue.

 

[Damir Halilovic]

ASUS Z87-C Motherboard
Processor Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz, 3401 Mhz, 4 Core(s), 8 Logical Processor(s)
BIOS: BIOS Version/Date American Megatrends Inc. 2103, 15.8.2014
I/O Port 0x000003C0-0x000003DF NVIDIA GeForce GTX 780
2x Kingston Predator 8GB DDR3 1866Mhz RAM
CoolerMaster 700W Power Supply

 

[Damir Halilovic]

So, I did a few more tests and now I think I need an exorcist, not tech support.

1. This morning, FurMark reliably caused the crash within 3 minutes at 1920x1080, max AA, fullscreen burn test.
2. After the 3rd or 4th crash I went into my BIOS to check my GPU voltage, AND it crashed while in my BIOS. This is surely a fault GPU I thought.
3. After the second BIOS crash (right after the first), it wouldn't boot up at all. My hard drives didn't even appear in the boot menu. This made me think it's a motherboard issue all of a sudden.
4. After fumbling with the SATA cables (didn't even remove them, just... well, wiggled them around), it booted up normally.
5. 30m later I run FurMark again to see what the power % at the time of the crash it. It's been running without a hitch ever since (45m at the time of writing). No crash at all. The temp peaks at 83 degrees celsius and power % howevers around 97%.

Now, I think all of this points to a faulty PSU. If the PSU is faulty I believe my GPU would be the first one to show symptoms as it drains most power, so any issue would cause it to kick the bucket first. The HDD absence would also be explained with that. I suppose it is possible that me cleaning the PC would have alleviated the problem temporarily due to the decreased system heat and the lower fan power consumption (as they aren't spinning full speed all the time). It would also explain why the issue is popping up so periodically, the voltage on the mains fluctuates in my country from time to time, this could explain it.

Still doesn't make sense that the last time I had it I fixed it with a chipset driver / BIOS update...

I will replace my PSU in the coming week and see if that does anything.