Is my GPU causing BSOD's? (Critical_Process_Died)

[azotevirtiol]

Hey,

I bought a RTX 2080 a few weeks ago and installed/ran it without any problems until about 3 days ago. My computer has been getting BSOD's of Critical Process Died at what feels like random. It's only ever when running games, however.

After trying to troubleshoot for a day, I gave up and formatted/reinstalled windows 10 from scratch - the issue persists. I believe it's a hardware problem.

The games it has crashed on so far have been 03.04 (a small unity horror game) and MH:W, as well as PUBG. The BSOD does not leave any kind of dump.

I have ran driver verifier and did not get any errors, as well as checked my disks and ran Window's RAM test with no problems.

Any help would be fantastic.

Edit: Forgot to mention; I installed CAM to monitor what was going on and I noticed that my GPU clocks were almost hitting 2100MHz, I have not overclocked this card and there are no auto-OC options as far as I'm aware of.

Edit 2: The BSOD causes Windows to load permanently on bootup. Not a bootloop (I think), but after the BIOS splash I just see the loading wheel infinitely. I have to restart the computer until windows re-installs itself to boot properly again.

My specs:
CPU: Ryzen 7 1800X
RAM: Corsair 16GB DDR4 Vengeance LPX 3200MHz
GPU: AORUS RTX 2080 8GB XTREME
PSU: Corsair RM750x Gold Certified 750W
MOBO: MSI Gaming Pro Carbon x370

 

[Barty1884]

GPU boost will take the card higher, depending on voltage/temp headroom..... but 2100MHz does seem a little high to me.

Given the hardware in question.. Personally, I'd start by disabling DOCP/XMP and run the memory at stock 2133/2400MHz and see if the issue persists.
While it would usually throw up different errors, it's a common place to start on a first Gen Ryzen chip/board.

While Windows RAM test might've based, it's by no means an exhaustive test. MemTest86 would be what I'd look to run to confirm full stability, but for now, a quick & dirty 2133/2400MHz should give you an idea to rule it in/out.

It's entirely possible the GPU is the culprit here, but I would be surprised, honestly.

 

[azotevirtiol]

GPU boost will take the card higher, depending on voltage/temp headroom..... but 2100MHz does seem a little high to me.

Given the hardware in question.. Personally, I'd start by disabling DOCP/XMP and run the memory at stock 2133/2400MHz and see if the issue persists.
While it would usually throw up different errors, it's a common place to start on a first Gen Ryzen chip/board.

While Windows RAM test might've based, it's by no means an exhaustive test. MemTest86 would be what I'd look to run to confirm full stability, but for now, a quick & dirty 2133/2400MHz should give you an idea to rule it in/out.

It's entirely possible the GPU is the culprit here, but I would be surprised, honestly.

I've set my speeds down to 2400MHz and disabled A-XMP. I'll report back if it still crashes me out. If it does I'll try and get memtest working, couldn't last time I tried because it wouldn't boot from my USB for some reason - terrible with BIOS settings. My assumption on the GPU being the problem was simply because it's the newest part. Everything else is roughly a year old.

 

[Barty1884]

What GPU did you have in there before? Did you fully remove any prior GPU drivers before installing the 2080?

While the memory may be ~12months old and appeared stable.... The GPU is likely rendering more frames &/or allowing you to play at higher settings which, in-turn, can ask more of your other system resources. It just might have been enough to show symptoms of unstable memory.

Not a guarantee, just a quick & relatively easy place to start on a Ryzen platform with no outright obvious issues.
Of course, if you didn't remove prior GPU drivers, that may well be the cause. If that's the case, run DDU in safemode to fully remove GPU drivers:
https://www.guru3d.com/files-details/display-driver-uninstaller-download.html

Then reboot/reinstall.

 

[azotevirtiol]

What GPU did you have in there before? Did you fully remove any prior GPU drivers before installing the 2080?

While the memory may be ~12months old and appeared stable.... The GPU is likely rendering more frames &/or allowing you to play at higher settings which, in-turn, can ask more of your other system resources. It just might have been enough to show symptoms of unstable memory.

Not a guarantee, just a quick & relatively easy place to start on a Ryzen platform with no outright obvious issues.
Of course, if you didn't remove prior GPU drivers, that may well be the cause. If that's the case, run DDU in safemode to fully remove GPU drivers:
https://www.guru3d.com/files-details/display-driver-uninstaller-download.html

Then reboot/reinstall.

I had a GTX 1060 (Aorus) before. I did run DDU, but I've also installed Windows10 on a fresh formatted SSD since then figuring it would be the simplest solution when the problem wasn't clear. The crash has happened on both stock and up-to-date drivers before and after the reinstall. Each crash has cause an infinite loading screen booting into W10 which I could only fix by rebooting 3x in a row and initiating the automatic troubleshooter that reinstalled W10 while wiping most of my files. This is why I do not have any crash dumps, making this much harder to fix.

I really appreciate the help, though. Currently downloading both MH:W and 03.04 to see if they work with the slower RAM speeds - MH:W will take ~4 hours, though. So I may not be able to report back on that until tomorrow night. Both games crashed out before the 20m mark.

 

[Barty1884]

Right! Sorry, you did say that, I forgot.

 

[azotevirtiol]

BSOD happened again. Roughly 15 minutes in. It was MUCH slower this time, though.

The game froze, sounds still played and after about 20 seconds the screen went black. Another 15-20 seconds after and I got the same BSOD. Every single time before this, it would BSOD instantly.

Though I did not need to force a repair this time - it rebooted normally which had only happened once out of the 10+ crashes over the past few days. Game performance was fine. Never below 80fps in 1440p.

Same as always - no crash dumps. Rebooted at 0%

 

[Barty1884]

I'll see if I can get some other opinions in here to help out.... I want to make sure you're able to rule out as much as possible before having to address the GPU.

 

[azotevirtiol]

I'm running memtest right now to make sure the RAM is fine. Changing it definitely had an affect. I'll report back once it's done, think it's set for 4 passes.

 

[Eximo]

There have been reports of bad Turing GPUs, more so than usual I would say. I thought it was mostly with the 2080Ti, but that might have been buyer's bias, anyone with a 1080Ti had little reason to go out and buy the RTX2080. With a fresh OS, and what I can only imagine is fine hardware otherwise, I suspect the GPU.

Possibly power delivery, if you haven't swapped some cables around and tried different ports on the PSU, give that a try. Could be a loose connection that fails as it heats up. Presumably your old card was only running a 6-pin, so you are certainly using the remaining +2 pins that hung off, or swapped cables completely.

I once troubleshot a computer for over an hour, bad PSU modular SATA cable shorting out the power. Oddly it still showed a boot screen, as soon as Windows attempted start, and presumably tried to turn on more things, it restarted.

 

[azotevirtiol]

I'll try swapping the cables if memtest doesn't show anything wrong! I only have 6+2 pin cables, but I have two of them. I'll swap them out and try another slot on the PSU. Though I have ran several 'stress tests' on my machine. Fur-something and two different CPU ones. They all showed fine. The machine seems to shut down around the 20 minute mark every single time - only exception being PUBG where I would BSOD, reboot, relaunch and then BSOD a few minutes later. That was the one instance where I didn't need to reinstall windows. I didn't launch the game a third time because by that point I was dead and we moved onto another game - Ring of Elysium - that didn't crash me in the 4 hours we played. I crashed in 03.04 17 minutes in-game straight after I got off Elysium.

 

[azotevirtiol]

Memtest finished overnight, 4 passes with no problems. I'll try the cables later in the day when I'm back from work.

 

[Barty1884]

Yeah, try the cables & if no change.... probably time to return or RMA that GPU.

 

[donalddeschamps]

I have the same issue after putting in an rtx 2070. Even after a clean windows was installed. My step son starts gaming and BAM...BSOD

 

[azotevirtiol]

I have the same issue after putting in an rtx 2070. Even after a clean windows was installed. My step son starts gaming and BAM...BSOD

Oh boy, I totally forgot to update this. Wanted to leave it until I was sure I'd fixed it.

The problem was actually my SSD. Disk checker tools claimed it was working fine, but after I removed it and installed Windows on my HDD there hasn't been a problem.

If you've got an SSD I'd try doing that. If you haven't, and everything else in the thread doesn't help I'm afraid I don't know.

 

[donalddeschamps]

I have the same issue after putting in an rtx 2070. Even after a clean windows was installed. My step son starts gaming and BAM...BSOD

Oh boy, I totally forgot to update this. Wanted to leave it until I was sure I'd fixed it.

The problem was actually my SSD. Disk checker tools claimed it was working fine, but after I removed it and installed Windows on my HDD there hasn't been a problem.

If you've got an SSD I'd try doing that. If you haven't, and everything else in the thread doesn't help I'm afraid I don't know.

So quick question. Is windows failing on the drive or the game. Hipothecaly if I had windows on a 250 GB hard drive and had games saved and installed on the 1 tb harddrive we assume is failing, would that fix the problems maybe?