Question Are these signs my GPU is dying?

[c0mplexMX]

Hey guys, hope y'all doing good.
So I've been having issues with my GPU for a while now, it "blacks out" sometimes while playing, the screen goes black and the fans start spinning 100% speed, but the PC stays on and I also still have audio and my friends can still hear me on discord, but there's no response to any input at all and I'm forced to restart the PC in order to use it again. The PC also sometimes freezes as soon as the Windows login screen shows up, the screen goes black and I have to restart. Yesterday, I was playing Apex Legends, the computer completely froze, the screen went black, started flashing gray-black-gray-black and then some kind of " TV white noise static" thing started to intermittently appear (

https://imgur.com/x8G1KND

View: https://imgur.com/x8G1KND
), then it just restarted by itself and right after I logged into Windows, these... artifacts? appeared and the PC completely froze again (

https://imgur.com/KjdwO8b

View: https://imgur.com/KjdwO8b
), had to manually restart.
Already tested on another PC, changed my own PSU and motherboard but the issues are still there, it's not temps or drivers either. Also tested RAM with memtest64, no errors at all. I also tried the stress test provided by AMD itself and it marks at least one error, memtestcl marks 0 errors and OCCT just doesn't post anything.
Could my GPU be dying? I still have valid warranty, but idk if I should send back the GPU or if this could be just a software problem, which I highly doubt.
The GPU is a little over 2 years old. I've experienced the same issue in other games such as Genshin Impact, Rocket League, Valorant, 7 Days to Die, even Minecraft with some shaders. I didn't send the GPU back yet bc the issue would go away for a couple weeks but eventually come back.

System specs:
CPU: Ryzen 5 2600
Mobo: B450 Aorus Elite V2
PSU: Corsair CX650M
RAM: XPG D60G 2x8GB@2933MHz
GPU: Gigabyte RX 590 Rev. 2.0 - No OC at all, ever.

 

[Lutfij]

Welcome to the forums, newcomer!

Do you get an error Code 43 in Device manager? If so, then yes the GPU's dead. You can cross reference this claim by using GPU-Z and seeing what your memory shows up as. it should be 0MB.

 

[c0mplexMX]

Welcome to the forums, newcomer!

Do you get an error Code 43 in Device manager? If so, then yes the GPU's dead. You can cross reference this claim by using GPU-Z and seeing what your memory shows up as. it should be 0MB.

Hi, and thanks!
Device manager says the device is working properly, and the memory shows up as 8192MB on GPU-Z.
There's a couple errors on event viewer but I'm assuming those are normal since they aren't logged at the time of the crashes?

 

[DRagor]

Already tested on another PC

Could you clarify this? What you tested on other PC and what was the result?
Another thing to check is your monitor and cable. This is not common but it could be them failing.

 

[c0mplexMX]

Could you clarify this? What you tested on other PC and what was the result?
Another thing to check is your monitor and cable. This is not common but it could be them failing.

I tested the GPU on a different PC, different monitor and everything, and it still blacked out and fans went 100% speed.
Then I changed my own mobo and psu, I had an A320 from MSI and a Corsair CX550M, but even though the problem went away for a few days, it came back

 

[DRagor]

I tested the GPU on a different PC, different monitor and everything, and it still blacked out and fans went 100% speed.
Then I changed my own mobo and psu, I had an A320 from MSI and a Corsair CX550M, but even though the problem went away for a few days, it came back

Ok. So in this case there are only two possibilities: GPU or drivers. If you have not done it yet uninstall current drivers with DDU and then reinstall. Also I agree with Lutfij it looks like VRAM problems, so check how GPU temps look like, particularly VRAM temp.

 

[c0mplexMX]

Ok. So in this case there are only two possibilities: GPU or drivers. If you have not done it yet uninstall current drivers with DDU and then reinstall. Also I agree with Lutfij it looks like VRAM problems, so check how GPU temps look like, particularly VRAM temp.

Already made a clean install of the drivers, tried using a version before the issues first appeared, but didn't work. I'm gonna check the VRAM temps

 

[c0mplexMX]

Ok. So in this case there are only two possibilities: GPU or drivers. If you have not done it yet uninstall current drivers with DDU and then reinstall. Also I agree with Lutfij it looks like VRAM problems, so check how GPU temps look like, particularly VRAM temp.

I couldn't find how to specifically monitor de VRAM temps, but the temps overall didn't get past 82 degrees

 

[DRagor]

I couldn't find how to specifically monitor de VRAM temps, but the temps overall didn't get past 82 degrees

Use GPU-Z. Obviously the most interesting moment is just before it crashes.

 

[c0mplexMX]

Use GPU-Z. Obviously the most interesting moment is just before it crashes.

Just came back from an Apex Legends session. It did crash but idk how to share the GPU-Z log, pastebin messes it up and is unreadable

 

[Third-Eye]

@c0mplexMX

RMA the GPU while you still have warranty.

Since it's doing the same thing in another PC, it's most definitely a faulty GPU that is only being revealed by the driver interacting with the hardware. The Microsoft display driver used in safe mode or before the full driver is installed, would not be interacting in the same way as the full graphics card driver in Windows.

 

[c0mplexMX]

@c0mplexMX

RMA the GPU while you still have warranty.

Since it's doing the same thing in another PC, it's most definitely a faulty GPU that is only being revealed by the driver interacting with the hardware. The Microsoft display driver used in safe mode or before the full driver is installed, would not be interacting in the same way as the full graphics card driver in Windows.

Yeah, I'm doing it tomorrow, not gonna wait until something that actually voids de warranty happens