• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Black screen in SOME games 6800xt Sapphire Pulse

mopi

New Member
Joined
Jun 14, 2024
Messages
7 (0.02/day)
When gaming in certain games after a while all my displays will turn black and nothing will happen (no audio as well). The power button no longer will work and the only way to turn it off and back on is to turn off the power supply manually. This sounds like it could be a psu issue, even tho a 750w psu should be enough for it from what I can tell) but I just want to be sure it's not the gpu because I got it 2nd hand and it only started happening after switching to this gpu...
 

Attachments

  • Navi 21.rom
    Navi 21.rom
    1 MB · Views: 59
  • 6800xt.gif
    6800xt.gif
    26 KB · Views: 130
When gaming in certain games after a while all my displays will turn black and nothing will happen (no audio as well). The power button no longer will work and the only way to turn it off and back on is to turn off the power supply manually. This sounds like it could be a psu issue, even tho a 750w psu should be enough for it from what I can tell) but I just want to be sure it's not the gpu because I got it 2nd hand and it only started happening after switching to this gpu...
It looks like a VRM issue. What happens when you put all the fans on 100%? Does it take longer to crash?
Can you tell us more about your system? Motherboard, processor, power supply (brand and model)
 
It looks like a VRM issue. What happens when you put all the fans on 100%? Does it take longer to crash?
Can you tell us more about your system? Motherboard, processor, power supply (brand and model)
Locking my GPU fans at 100% didn't seem to do anything, It still happened in about the same amount of time in this case.

Motherboard: B450 Tomahawk MAX
CPU: 5800X3D
PSU: Cooler Master MasterWatt 750W 80 Plus Bronze (MPX-7501-AMAAB-US)

One thing I thought I should point out is that GPU-Z reports that my bios version is 020.001.000.046.000000 but that doesnt seem to match what is listed on here for this gpu. However it does still pickup the correct name and stuff so I'm not sure...

Another weird thing is that it doesnt happen in every game, even some demanding games like MW3 have never ever had a crash like this in (some direct x crashes but thats a problem everyone has with that game)
 
Last edited:
I'd probably try a high quality's psu 850w or more from a seller with a good return policy like Amazon.
 
I personally had this issue with my XFX 6800XT. MOSFETs were dying on the card itself and ended up being tested on a 3rd desktop where it was confirmed dead.

Another thing I've seen is with my 6950XT, where the PSU couldn't handle the transients. These are the two parts you need to be testing. GPU, PSU. Try with other builds if you can.
 
Locking my GPU fans at 100% didn't seem to do anything, It still happened in about the same amount of time in this case.

Motherboard: B450 Tomahawk MAX
CPU: 5800X3D
PSU: Cooler Master MasterWatt 750W 80 Plus Bronze (MPX-7501-AMAAB-US)
I own a MSI B450 Tomahawk MAX motherboard before and had the exact same issue. The VRM on this motherboard are bad quality and is almost broken.
Due the higher temperatures in your case caused by the new graphics card your system crashes.

A workaround for now to disable disable PBO or put the processor in 65W eco mode. Do you have a lower end AM4 CPU which you can try?
Maybe even put the case on the side with the side panel removed so the hot air can easly move out of the case.

The only real fix is replacing the motherboard. With the suggestions I made before you should be able to tell if it is really the motherboard.

See my posts about this board :laugh:
 
I own a MSI B450 Tomahawk MAX motherboard before and had the exact same issue. The VRM on this motherboard are bad quality and is almost broken.
Due the higher temperatures in your case caused by the new graphics card your system crashes.

A workaround for now to disable disable PBO or put the processor in 65W eco mode. Do you have a lower end AM4 CPU which you can try?
Maybe even put the case on the side with the side panel removed so the hot air can easly move out of the case.

The only real fix is replacing the motherboard. With the suggestions I made before you should be able to tell if it is really the motherboard.

See my posts about this board :laugh:
I disabled PBO in the bios and have had no crashing issues throughout the day playing stuff that would normally cause a crash. So I'm guessing this was in-fact the issue, thank you so much for saving me a big headache trying to diagnose this.

Edit: just after posting this I did end up crashing for the first time today, but it was massively improved from disabling PBO so I think the motherboard could still be the cause
 
Last edited:
RDNA2 can trip OCP on a high end 850W PSU.
if your hotspot isn't att 113°C non stop before it happens, i'd change the PSU.
 
I disabled PBO in the bios and have had no crashing issues throughout the day playing stuff that would normally cause a crash. So I'm guessing this was in-fact the issue, thank you so much for saving me a big headache trying to diagnose this.

Edit: just after posting this I did end up crashing for the first time today, but it was massively improved from disabling PBO so I think the motherboard could still be the cause
It is possible to narrow it further down while PBO is disabled. You can block the air intake of the case with towels in front so less air can be pulled in. Temperatures will rise.
If the systems crashes more, then you know it is caused by heat and not OCP of the PSU.

Then the question is, which VRM is broken? One on the motherboard or the graphics card. I think its the motherboard due my experience with the same board......
 
If anyone is up to it I decided to log while I play ff14 (a game that frequently crashes) with hwinfo, its a bit over an hour long before I crash. I honestly dont know what I would be looking for here for an abnormality. One this that sticks out to me is that the VR MOS temps are recorded really weirdly (only reports at full degree intervals, seems really low temp, and doesn't move much at all) but they jump to what is reported as 39 degrees right before I crashed.

I haven't mentioned it before but theres no log in windows event viewer of the crash besides kernel-power.
 

Attachments

Last edited:
If anyone is up to it I decided to log while I play ff14 (a game that frequently crashes) with hwinfo, its a bit over an hour long before I crash. I honestly dont know what I would be looking for here for an abnormality. One this that sticks out to me is that the VR MOS temps are recorded really weirdly (only reports at full degree intervals, seems really low temp, and doesn't move much at all) but they jump to what is reported as 39 degrees right before I crashed.

I haven't mentioned it before but theres no log in windows event viewer of the crash besides kernel-power.
Does only FF14 crash or the whole computer hangs on a blackscreen and the power button is not responding?
 
Does only FF14 crash or the whole computer hangs on a blackscreen and the power button is not responding?
The whole computer hangs on a blackscreen and the power button is not responding
 
The whole computer hangs on a blackscreen and the power button is not responding
The VRM temperature of the motherboard is quite low. I dont expect this kind of crash at a VRM temperature of 39 degrees unless the VRM is broken to a level it can no longer deliver power at 39 degrees.

Do you know someone nearby with a high end AM4 system which you can ask if you can test the power supply and graphics card?
Then we know if one of those parts is faulty (GPU) or insufficient (PSU).
If it doesn't crash, then the culprit should be the motherboard but with these temperatures I have my doubts.
 
disabling xmp seems to fix the issue whatever it is, idk if this means its the ram, mobo, psu still lol
 
disabling xmp seems to fix the issue whatever it is, idk if this means its the ram, mobo, psu still lol
If you follow the same logic, then it can also be the processor :rolleyes: Disabling PBO and xmp lowered the energy requirement and lowering the performance.
Disabling PBO and xmp are reducing the stress on the processor and therefore also the vrm. My intuition says it is the motherboard....
 
Update for anyone curious, the issue was indeed the power supply. I bought a fsp hydro g 1000w from amazon to test it out and all my problems are gone, re-enabling pbo and xmp, everything is working perfectly.
 
Great to hear your problem is solved :D
 
I own a MSI B450 Tomahawk MAX motherboard before and had the exact same issue. The VRM on this motherboard are bad quality and is almost broken.
Due the higher temperatures in your case caused by the new graphics card your system crashes.

A workaround for now to disable disable PBO or put the processor in 65W eco mode. Do you have a lower end AM4 CPU which you can try?
Maybe even put the case on the side with the side panel removed so the hot air can easly move out of the case.

The only real fix is replacing the motherboard. With the suggestions I made before you should be able to tell if it is really the motherboard.

See my posts about this board :laugh:
I haven't been keen on msi for a very long time myself, same with gigabyte, Asus is on that list now too
 
I haven't been keen on msi for a very long time myself, same with gigabyte, Asus is on that list now too

With how long you've seemingly been at this I'm surprised every company isn't on your $#!+ list.......

I've had my good and terrible experience with all of them lol.
 
With how long you've seemingly been at this I'm surprised every company isn't on your $#!+ list.......

I've had my good and terrible experience with all of them lol.
AM3 Asus TUF 990FX, when AM4 came about TUF is abysmal, AsRock Steel Legend Replaced them for AM4.
 
Back
Top