GPU operates adequately for some time before it stops functioning properly.
GPU operates adequately for some time before it stops functioning properly.
I've been using my RX 580 for about four years now, and the overall system remains unchanged in age. There are only minor issues like occasional AMD driver malfunctions, no graphical errors, and no visual artifacts that suggest the GPU is failing. Recently, while playing Fallout 4, my GPU suddenly stopped transmitting a signal without warning. I could still hear the game and communicate with my Discord friend, but the monitor displayed a "no signal" alert. I attempted to restart the PC, yet the GPU continued to refuse any output. After leaving it overnight, it worked briefly again. Through some testing, I realized the RX 580 loses signal once it hits a specific temperature threshold. It comes with a 0 RPM setting, which means when the GPU cools down after being off for some time, everything functions normally. However, once the 0 RPM state is reached and the fans kick in, the signal drops instantly.
What’s more unusual is that even in quiet mode, the GPU still loses signal—slower than usual due to its cooling curve—but it can still handle 3D tasks without issues. I tested Cyberpunk and a modded version of Skyrim, and there were no graphical problems. Frame rates stayed steady, loading times remained consistent, and the GPU responded well to ADR. It even detected me without any trouble.
I’ve ruled out a power supply issue by using a spare GPU and a different PSU, which still caused signal loss. I also checked the thermal paste, but it didn’t help. The problem seems to be more related to power delivery. Using GPT, I suspect a capacitor failure inside the GPU when it tries to draw more power. I opened it up and inspected all the capacitors—none showed signs of leakage, burnout, or discoloration.
I’m wondering if anyone else has faced similar problems. Is this a permanent issue? Are there any steps I could take to revive it? It’s really frustrating not being able to figure out what’s happening, and I’ve read that GPUs rarely fail like this unexpectedly. I don’t have the budget for a replacement right now, so any advice or help to get it working again would be greatly appreciated.
Welcome to the forums, newcomer!
When posting a thread of troubleshooting nature, it's customary to include your full system's specs. Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
include the age of the PSU apart from it's make and model. BIOS version for your motherboard at this moment of time.
I've already tried changing the thermal paste and that did not have any results, though the GPU was not overheating in the first place.
What thermal paste did you use to replace the one applied from the factory? Did you replace the thermal pads as well?
It's possible that the GPU is on it's way out. The RX580 was a hot headed GPU to begin with.
I've read that even if it can happen, GPUs don't just die out of the blue like this too often.
But they do happen. You could try and reflash the VBIOS on your card and maybe that would help resolve the issue but there's a high chance you might end up with a paperweight if the process goes South.
I didn't swap out the pads, but temperature wasn't the main issue. I kept tracking both temperatures and voltages on the PSU and the 580 before any signal losses occurred. The 12V rail stayed above ~11.930, and temperatures never went past around 60°C during the crash.
I also refreshed the post with all the system details I have now.
I'm not certain about the precise model. It appears to be an older Acer 75Hz TN monitor featuring one of the longer number names. I haven't tested it myself, but I can certainly offer an explanation if you'd like.
This situation might actually be due to the monitor aging if the RX 580 is set up with a larger screen or higher refresh rate. However, if it relates to the fan, the problem is probably caused by the fan itself rather than power supply issues. It seems the fan might require more power than it was designed for to start, so cleaning and lubricating it could help. If it's not a brushless fan, try spinning it by hand to check for any breakage, using non-conductive grease. You might find the model number on the fan’s back, and you can search online to locate a new one for a reasonable price.
On motors with brushes, friction gradually reduces brush length and stops contact. If this applies to your fan, consider replacing the brushes instead of buying a new motor. Computer fans are often delicate and small, requiring special tools or techniques to open them without damaging the axle.
Display Driver Uninstaller (DDU) version 18.1.4.1 is available for download
Use DDU to delete and reinstall graphics drivers cleanly, ensuring no remnants remain, such as registry entries. Visit www.guru3d.com for more details.