Pc crash / black screen reset

[Acuity]

Hi guys, I'm writing here because I think you are the best forum in .
I'm having a problem when I use my PC with heavy games, often games in UE5.
After playing Remnant 2 for about half an hour for example, my PC shows a black screen and then crashes and resets.
I tried several applications that could test the stability of the hardware and I didn't encounter any problems. Then I wanted to try with OCCT and the PC passed all the tests for an hour, except one. The VRAM one, in that test the PC had the exact same problem as when I play Remnant 2, black screen and hard reset. Exactly after a few seconds, I attach the video of when I do the test with hwinfo overlay, do you experts notice anything strange in the GPU data from the video?
I have already changed the PSU and bought a brand new TUF 850w ATX 3.0 and the problem occurs the same so I exclude the PSU.
Thanks to all guys.
P.s I used the search function of the forum and I noticed others with my same problem (critic kernel power 41) and some of them solved it by simply putting a support for the gpu and inserting it better, which I have already tried but nothing changed.

The photo is from the moment before the crash

PC spec:
Tufb650plus
ddr5 silicon power 6000 cl 30
1tb p5 plus crucial
psu 850tuf atx 3.0
GPU rx 6800 xfx speedster
CPU ryzen 5 7600

 

[VuurVOS]

Hi guys, I'm writing here because I think you are the best forum in .
I'm having a problem when I use my PC with heavy games, often games in UE5.
After playing Remnant 2 for about half an hour for example, my PC shows a black screen and then crashes and resets.
I tried several applications that could test the stability of the hardware and I didn't encounter any problems. Then I wanted to try with OCCT and the PC passed all the tests for an hour, except one. The VRAM one, in that test the PC had the exact same problem as when I play Remnant 2, black screen and hard reset. Exactly after a few seconds, I attach the video of when I do the test with hwinfo overlay, do you experts notice anything strange in the GPU data from the video?
I have already changed the PSU and bought a brand new TUF 850w ATX 3.0 and the problem occurs the same so I exclude the PSU.
Thanks to all guys.
P.s I used the search function of the forum and I noticed others with my same problem (critic kernel power 41) and some of them solved it by simply putting a support for the gpu and inserting it better, which I have already tried but nothing changed.

The photo is from the moment before the crash

PC spec:
Tufb650plus
ddr5 silicon power 6000 cl 30
1tb p5 plus crucial
psu 850tuf atx 3.0
GPU rx 6800 xfx speedster
CPU ryzen 5 7600

Can you test the graphics card in a different system?

 

[eidairaman1]

Reseat everything, update your motherboard bios

 

[Acuity]

Can you test the graphics card in a different system?

Unfortunately I can't, I only have one PC and none of my friends have this passion, so they don't use gaming PCs

 

[Acuity]

Reseat everything, update your motherboard bios

I updated the bios to version 3034 and 3035 (the latest) unfortunately nothing has changed. I then proceeded to completely format each Windows 11 disk and downloaded only steam and remnant 2, nothing else. It crashed after 1 hour in the same way

 

[eidairaman1]

I updated the bios to version 3034 and 3035 (the latest) unfortunately nothing has changed. I then proceeded to completely format each Windows 11 disk and downloaded only steam and remnant 2, nothing else. It crashed after 1 hour in the same way

Any shops nearby? You are running out of troubleshooting options

 

[Acuity]

Any shops nearby? You are running out of troubleshooting options

Yes I think I've tried everything. Tomorrow I'll take a look in some nearby shop. Anyway it only happens to me with UE5 games and it also happened to me in cyberpunk, but only when I activated RT ultra. So when the gpu is working very intensely. For example in light games, like FC24 I tried keeping it on all afternoon and the black screen and reset never came out.
Anyway thanks for your time guys

 

[Acuity]

I'm updating the thread for those who may come by later with the search function because they have the same problem as me.
After trying in every way to solve the black screen and hard reset of the PC while playing graphically demanding games, it seems that I managed to solve the problem by undervolting the RX 6800 -30 offset from the adrenaline panel and loosening the screws of the AIO. I don't know which of the two was the solution, but I solved it this way for now, I thought it was a driver problem but I tried them all and it's not a driver problem, not even a psu problem.
(I had read about loosening the screws of the AIO on the AMD website to be exact) I did a 4-hour session without any crashes and it had never happened to me.
However, I think it's more of a GPU voltage problem, but tomorrow I'll try to remove the undervolt and see if the solution was instead the AIO.

 

[Caring1]

KP 41 tend to be be an overheating issue.

 

[eidairaman1]

KP 41 tend to be be an overheating issue.

Which if a pcb flexes causes concaveness to the gpu die.

Nowadays I wouldn't recommend thermal pastes for gpu dies but a thermalpad.

 

[Acuity]

KP 41 tend to be be an overheating issue.

Strange because my GPU temperatures are around 75 core and 85 hotspots, while the CPU is 65. Not so bad right?

 

[eidairaman1]

Strange because my GPU temperatures are around 75 core and 85 hotspots, while the CPU is 65. Not so bad right?

Yes however if your pcb is warped it can cause bad contact leaving an airgpap