[SOLVED] RX 590 Black Screen/Fans Spin Up/PC Still Running

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
OS: Windows 10 Pro 20H2
CPU: Intel i5 9400F
GPU: XFX RX 590 FATBOY
RAM: GSKILL 16GB DDR4
MoBo: Gigabyte Z390 UD
PSU: Raidmax RX 630SS

I built this PC at the end of last year and up until the WoW Shadowlands PreExp Patch I never had any issues playing any games @ Ultra settings. Out of no where, while playing WoW Shadowlands (can be fishing or in a heated battle) my screen will just go black and say "No Input Detected". The fans in my computer will spin up, the PC will stay running and I can even hear the sound still going....

At first, this would only happen every once in a while. I thought maybe it was an issue between the newer version of the game and my drivers so with every driver update I was quick to install it. To my surprise though, with every update my issue would happen more and more often. I read that replacing the thermal paste would be a highly likely remedy and while it seemed to be a fix for the time being it still never went away.

The "ONLY" thing that seems to keep this issue from happening is limiting the in game FPS to 50-58.... In other games and with the limiter off I get 120-200+ FPS so this fix seems wrong in many ways... I have watched the Temp in hardware monitor and I have not seen it pass 83C Under Full load. Please! Any help would be appreciated!

***Side note: After I installed the new thermal paste and for a while before I updated the drivers again the issue changed. The screen would freeze for 20 secs and then an AMD message would pop up saying that they detected a timeout in the drivers. I assumed that was throwing the GPU into whatever it was doing and that my changing the thermal paste fixed part of it but after a couple driver updates the black screen was back instead....
 
Last edited:
Solution
This is just a link to one of the Fatboy's bios files: https://www.techpowerup.com/vgabios/205256/xfx-rx590-8192-181031

Driver timeouts can be linked to bad gpu overclocks and corrupted/bad drivers. Have you tried the Display Driver Uninstaller(requires Safe Mode to run) route yet?
The gpu disappearing from the face of the Earth is not a very good sign...

That particular model had been shown to have a weak cooler, and if it's been running toasty all this time, the card, or parts of it, may have been damaged via thermal expansion... but I can't rule out the possibility the psu couldn't have done something either.
Some models are silent killers.

Phaaze88

Titan
Ambassador
I built this PC at the end of last year and up until the WoW Shadowlands PreExp Patch I never had any issues playing any games @ Ultra settings. Out of no where, while playing WoW Shadowlands (can be fishing or in a heated battle) my screen will just go black and say "No Input Detected". The fans in my computer will spin up, the PC will stay running and I can even hear the sound still going....
Straight up sounds like power supply - supplied voltage dropping too low.

I read that replacing the thermal paste would be a highly likely remedy...
The gpu would have to be overheating. Were you monitoring thermals before you did this fix?

The "ONLY" thing that seems to keep this issue from happening is limiting the in game FPS to 50-58.... In other games and with the limiter off I get 120-200+ FPS so this fix seems wrong in many ways... I have watched the Temp in hardware monitor and I have not seen it pass 83C Under Full load.
Not capping the fps, or setting a higher cap than what is set currently makes the gpu run hotter and pull more power... if the psu can't keep up with the higher gpu loads - aka, the voltage drops...


***Side note: After I installed the new thermal paste and for a while before I updated the drivers again the issue changed. The screen would freeze for 20 secs and then an AMD message would pop up saying that they detected a timeout in the drivers. I assumed that was throwing the GPU into whatever it was doing and that my changing the thermal paste fixed part of it but after a couple driver updates the black screen was back instead....
Not sure what happened here, but before this, most of everything else was pointing towards the psu.
 

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
This is the main message I was getting when I didnt get a full Black screen (No Input Found).
https://ibb.co/vwLnzf3

I then started getting this one also...
https://ibb.co/QYnjsVr

I decided to grab a new PSU and went with the EVGA 600 GD 100-GD-0600-V1. All day I had no issues where otherwise I did. I left games open while I went about the house doing other things. At the very end of the day.... Just before I went to sleep... I get on the game just to check something simple and Boom... Black screen...

According to the Error in the second picture (From Overwatch) it says it can be caused by OC'ing... The only thing I have on in the Bios is the RAM XMP Profile. I turned that off to see if that is my next step but do you think that could be causing this? My other question right now, this and my previous PSU has the 4+4 CPU power for the MoBo and I have that plugged into the 8pin on the board... Is this right? or Should I disconnect that and just plug one into the 4pin? Could that be where the issue is?
 

iPeekYou

Distinguished
Jul 7, 2014
392
77
18,790
According to the Error in the second picture (From Overwatch) it says it can be caused by OC'ing... The only thing I have on in the Bios is the RAM XMP Profile. I turned that off to see if that is my next step but do you think that could be causing this? My other question right now, this and my previous PSU has the 4+4 CPU power for the MoBo and I have that plugged into the 8pin on the board... Is this right? or Should I disconnect that and just plug one into the 4pin? Could that be where the issue is?

4+4 is alright, it's meant to do the same as a straight 8 pin. ATX standard used to call for 4 pin CPU power, but later PSUs started adopting the EPS standard from server spec with 8 pin power.

83°C is awfully close to 85°C, which from my experience with 580 is the thermal limit. Mine used to do 82-83°C under max load as well, but I sold it before I faced any issues.

My guess is thermals. You mentioned you had some progress with it, might worth pursuing. Check if the cooler is remounted properly and maybe try underclocking/undervolting a bit. Or use Radeon Chill to do that.
 

Phaaze88

Titan
Ambassador
This is just a link to one of the Fatboy's bios files: https://www.techpowerup.com/vgabios/205256/xfx-rx590-8192-181031

Driver timeouts can be linked to bad gpu overclocks and corrupted/bad drivers. Have you tried the Display Driver Uninstaller(requires Safe Mode to run) route yet?
The gpu disappearing from the face of the Earth is not a very good sign...

That particular model had been shown to have a weak cooler, and if it's been running toasty all this time, the card, or parts of it, may have been damaged via thermal expansion... but I can't rule out the possibility the psu couldn't have done something either.
Some models are silent killers.
 
  • Like
Reactions: spentshells
Solution
if you bought the 590 last year it could have been sitting for for 3 years if you bought it new or if you got it used it might have been heavily used.
your temps are super high
remove the heatsink and clean the old paste and apply some new stuff after cleaning the heatsink of dust.
 

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
I bought it new. I also have already replaced the thermal paste. I am wondering though now that it isn't happening too much with the new PSU if I should do it again. Maybe because it was getting hot with the under voltage from the last PSU the paste dried some? IDK its worth a try right?

I noticed on my other PCs that when the GPU gets up into the 70s that the fan will kick up into 3k RPM. On this 590 even when it is reaching the 80s, the fan will not push past 2250 RPM. So maybe when it is able to just heat enough past what the 2250 RPM can cool is when it happens....? What is keeping it from kicking up higher? Seems like it has to black screen for the fans to go higher at stock right now.

I found a setting in the AMD Drivers under Performance>Tuning called "Auto Tuning" and the Option is "GPU Overclocked". It looks like it does a small OC to the card but the main thing I noticed is that the fan will ramp up to over 4k RPM under load. So far no issues, we will see how it goes from here.
 
Last edited:
  • Like
Reactions: spentshells

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
I have 6 case fans and its pretty cool inside there. Just stopping in to add a note that I found another thread somewhere that people suggested to turn off the "0 RPM" option. I tried doing this and as soon as I launched a game I got Particle squares all over the screen and then the PC Blue Screened. I went back to using the "GPU Overclocked" setting right away.
 

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
With the new PSU and/or the Setting in the Display Drivers I have gone 2 days with no Freezing, Errors or Black Screens. I feel as though the Fix was a combination of the Better PSU and the Driver Setting Forcing the GPU fan to do a better job at cooling the GPU. I will post back in a few more days with how it is going. Below is where the setting is found in the Driver Menu and My current GPU Monitor Stats under Full Load. I want to note that even though I chose to use this slight OC setting, I really feel the only thing it does is force the fans to run at Higher RPMs....
Driver Setting
https://ibb.co/hsWPkx0

HWMonitor
https://ibb.co/7kfGwW1
 

ToolPunkXS

Commendable
May 10, 2021
13
1
1,515
I can't believe it's been almost a year since I said I will post back in a few days... Well the issue came and went with driver updates and fiddling with the driver setting and I got fed up... So I tore it open again and applied MX-4 this time (first time all I had was Arctic Silver). The problem went away for several months and then slowly started creeping back and with every driver update it would get worse... I found out that I actually still have a warranty but was reluctant to send it in with the prices of GPUs right now... I did everything I could at this point though so I just sent it in and am awaiting a replacement... Let's hope the new one is free of this nightmare...