My screen keeps freezing and goes black for a few seconds

[Dot-_-Com]

My screen randomly freezes then goes black for a few seconds 

When my screen comes back on radeon notification pops up and says 

Default Radeon WattMan settings have been restored due to unexpected system

It happens a lot! it happens when i browse the internet. I cant even open up games anymore the game just crashes.
My system 
CPU -- Intel Core i5‑6500 3.2 GHz Quad‑Core Processor SkyLake

GPU -- MSI Radeon RX 480 4GB GAMING X Video Card

RAM -- G.Skill Ripjaws V series 16GB

SSD -- PNY 240GB

PSU -- SeaSonic 520W 80+ Bronze Certified Fully-Modular ATX Power Supply

Mother Board -- ASUS B150 GAMING/AURA

[DJ46]

Have you tried reinstalling the drivers/installing newer ones/rolling back to an older one?

 

(Supposedly just selecting clean install in the installer menu is better than other solutions - like DDU - for a fresh driver install.)

[Dot-_-Com]

Just now, DJ46 said:

Have you tried reinstalling the drivers/installing newer ones/rolling back to an older one?

 

(Supposedly just selecting clean install in the installer menu is better than other solutions for a fresh driver install.)

Yep I did a clean install of drivers and still doing it 

[DJ46]

Have you applied any overclocks/changed any settings in Wattman?

[Dot-_-Com]

1 minute ago, DJ46 said:

Have you applied any overclocks/changed any settings in Wattman?

No everything is default. Could it be my GPU is dying I need to replace it? 

[DJ46]

2 minutes ago, lillozzer68 said:

No everything is default. Could it be my GPU is dying I need to replace it? 

Maybe. What are the idle GPU temps?

 

Rather curiously, I found a reddit thread where a user with these exact issues ended up having to replace his RAM. Maybe you could try running memtest to check that?

 

(I'll try to respond tomorrow if necessary.)

[YaBoiWill]

I have the same sort of thing, the screen occasionally goes black, but I don't have the errors. Its on a rx580, and when in games/high system usage the core and memory clock goes really low, even though temps are low. 

Doing some napkin maths iv deduced that im drawling too much power, with hard drives, fans, cpu and gpu etc. 

I have a new psu coming that should have enough head room, so im going to see if that helps the situation. 

[Dot-_-Com]

3 hours ago, DJ46 said:

Maybe. What are the idle GPU temps?

 

Rather curiously, I found a reddit thread where a user with these exact issues ended up having to replace his RAM. Maybe you could try running memtest to check that?

 

(I'll try to respond tomorrow if necessary.)

Idle gpu temp 80f 

 

Memtest results - all ram sticks are good 

[DJ46]

13 hours ago, lillozzer68 said:

Idle gpu temp 80f 

 

Memtest results - all ram sticks are good 

Well I'll be damned. My PC did something similar yesterday, but without the radeon notification popping up. It happened twice so far, during a light load soon after booting up. Otherwise it seems fine.

I know I had the cooler seated incorrectly and ran a slight load for a second. The GPU temps went immediately to 88°C and I turned it off and shut the system down.

I think something might've gotten damaged. That would mean your card is damaged even more. I hope that's not the right answer to any of us

 

Your issue is definitely more limiting. I found some threads mentioning issues with freshly installed new drivers - maybe you could try to roll back to even older ones.

Other information points to a bad overclock. Your GPU might be factory overclocked and it might've become unstable due to aging silicon. You could try bumping the voltage up slightly if your temps are ok or lowering the core clock.

The last option I found so far is a registry tweak. I'm not sure what it's supposed to fix, it apparently forces you to restart your computer after a driver crash instead of trying to recover. (I really don't know how that could be helpful, I might look into it some more later.)

[Dot-_-Com]

1 hour ago, DJ46 said:

Well I'll be damned. My PC did something similar yesterday, but without the radeon notification popping up. It happened twice so far, during a light load soon after booting up. Otherwise it seems fine.

I know I had the cooler seated incorrectly and ran a slight load for a second. The GPU temps went immediately to 88°C and I turned it off and shut the system down.

I think something might've gotten damaged. That would mean your card is damaged even more. I hope that's not the right answer to any of us

 

Your issue is definitely more limiting. I found some threads mentioning issues with freshly installed new drivers - maybe you could try to roll back to even older ones.

Other information points to a bad overclock. Your GPU might be factory overclocked and it might've become unstable due to aging silicon. You could try bumping the voltage up slightly if your temps are ok or lowering the core clock.

The last option I found so far is a registry tweak. I'm not sure what it's supposed to fix, it apparently forces you to restart your computer after a driver crash instead of trying to recover. (I really don't know how that could be helpful, I might look into it some more later.)

WOW I hope your not going through what I’m going through It sucks. I think I should buy a new card and might as well upgrade to a 1070 ti 

[DJ46]

2 minutes ago, lillozzer68 said:

WOW I hope your not going through what I’m going through It sucks. I think I should buy a new card and might as well upgrade to a 1070 ti 

It seems to be lot less severe than your issue. I can game just fine.

 

Still, I would try some old drivers and underclocking the GPU before chucking it in a bin.

[Dot-_-Com]

46 minutes ago, DJ46 said:

It seems to be lot less severe than your issue. I can game just fine.

 

Still, I would try some old drivers and underclocking the GPU before chucking it in a bin.

It has never been overclocked and I have completely reinstalled drivers I don’t know what else to do 

[DJ46]

1 hour ago, lillozzer68 said:

It has never been overclocked and I have completely reinstalled drivers I don’t know what else to do 

Underclock it/give it slightly more voltage/both.

 

If underclocking it to a stupid level (like 800 core clock and 1200 memory clock or something, maybe even lower) doesn't help, it still might be just software.

[Dot-_-Com]

4 hours ago, DJ46 said:

Underclock it/give it slightly more voltage/both.

 

If underclocking it to a stupid level (like 800 core clock and 1200 memory clock or something, maybe even lower) doesn't help, it still might be just software.

I stressed test my cpu everything’s good 

I benchmarked my gpu it ran for half of a second and it frozen and screen went black 

I ran memtest again and everything’s fine. I think my gpu is the issue