Display crash (and signal lost) when playing games with high settings. Windows Event Viewer shows nothing

---- PC SPEC ----
CPU: Intel i5 4460 (1 year old)
GPU: MSI R9 390 (2 years old)
RAM: 16GB (8 + 8) (3 + 1 years old)
PSU: Seasonic S12II 620w 80+ Bronze (2 years old)
HDD: 1 1TB + 1 2TB (3 + 1 years old)
SDD: 128GB (1 year old)

OS: Windows 10
Monitor: A 1080p and a 720p
Current Radeon Driver: 18.3.2 (Adrenalin)

---

My PC has been running fine for few years. I did some upgrading here and there and this current build has been working fine for at least 1 years. So since a few weeks back, my PC keeps on crashing when I played max settings on DotA 2. Usually, it started with a flickering and few graphical artifacts (like colored squared) and then the monitor just turned black (both monitors display "No Signal"). The PC still runs (I can hear songs from Spotify). It didn't happen all the time but maybe 1 in 10 games. I thought it's just a GPU driver problem so I update the driver but it still happened. I tune the graphics settings down (into the 2nd lowest settings) and it didn't crash for 2 weeks now. Recently, I bought FFXV Windows Edition. Whenever I get into the first scene of the game (where Noctis fight Ifrit), the game starts to flicker, artifact appears and crash. I thought it was a random thing, so I reboot my PC, start the game again and the same thing happened.

I checked the log in Windows Event Viewer but weirdly there's no error message related to the crash. The only error message that I have is that:

"The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."

which is obviously because I force it to shut down. I also have this error message:

"The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID"

but I don't think it's relevant tho? Note that I have Speccy run when I play FFXV and the temperature are fine (GPU below 80 degree celcius and CPU below 70 degree celcius)

After that, I ran a stress test on my CPU with prime95 for like 10 minutes before my CPU tempt went up to 85 degree celcius. I had to stop the test since I don't want to take any risk damaging my CPU. Note that the test shows no error. I also did a ram test with Windows Memory Diagnostic and also shows no error. I then ran a stress test on my GPU with Furmark. Few minutes after its start (I think it's less than 2 minutes), the crash happened again. So I quickly force shutdown, reboot and check into Event Viewer and again, no error message.

My initial guess is it's a GPU issue. Since my GPU has another 1 year warranty, I plan to RMA it soon but my friend said that it could be PSU also since there's no crash log. I don't want to quickly make any decision to replace any of the hardware so I really need some help and opinion on this matter.

Thanks

Edit: Another thing to note is that my PC had been infected with bitcoin miner virus a few months back for few weeks before I realized it. Windows Anti-Virus didn't detect anything. For few weeks my GPU fan spins very fast and the temperature is quite high. Luckily I downloaded Malwarebytes and it manages to find it within seconds of scanning. However, this happened in 2016 if I remember correctly and my PC runs fine for the whole of 2017. Also, my desktop is very dusty.

Edit 2: I manually set the fan speed to 100% using MSI Afterburner. I noticed the idle temperature went down to 55 Celcius from 65 Celcius. I tried FFXV again. This time I can play longer than I did previously. I noticed that the temperature went up gradually and went it hit 75 Celcius, the game crashed BUT everything else still working (meaning I don't have to reboot the PC)

Submitted March 17, 2018 at 07:34PM by nyamuk91 http://ift.tt/2HIYtZh