Discussion AMD driver timeout: "Default Radeon Wattman settings have been restored." and "Unreal engine is exiting due to d3d device being lost error."

Jan 21, 2023
5
1
10
PC SPECS:
FX-8320
RX 580 ddr5 8GB
16 RAM
SSD Kingston 480gb
Redragon 500W 80 plus bronze (https://www.redragonzone.com/products/redragon-rgps-gc-ps001-500w-gaming-pc-power-supply)

GPU clock speed in games:
1250 MHz
VRAM speed:
2000 MHz
Temperature: 60° C

Crashes are random. I can play for hours before my games crash to desktop with the message error "Unreal Engine is exiting due to D3D device being lost" and AMD Adrenaline resets all video settings to default (also known as the Wattman crash error or driver timeout).

I tried different methods to fix the problem but none of them worked. Here's what I did:
  • I formatted my PC and did a clean install of my OS.
  • Reinstalled my GPU drivers.
  • Tried different versions of AMD Adrenaline. 20.11.2 / 22.3.1 / 22.11.2
  • Changed energy settings to High Performance in Power Management via the Control Panel. Turned off Fast Start Up, Sleep, Hibernate and Lock mode.
  • Disabled AMD Crash Defender via Device Manager and msconfig.
  • Ran the command sfc/scannow to scan the system and no files are corrupted.
  • Changed the TDR delay value to 10 or 60 decimal through regedit.
I don't know what else to do.

Some people said they had to undervolt their GPU to stop their games from crashing. I don't want to mess around with the GPU. It's still under warranty.
 
Last edited:
Can’t find a lot of info on that power supply. If you can, return it for a higher quality unit.


Take a look here, I don’t even see that unit on the list. Usually in that list you want to be at tier A or B.

A power supply being new has nothing to do with quality. I’ve used cooler master power supplies that you would start up a pc and the minute you started doing too much the computer would power off. Even without having a graphics card installed. Recommend to replace the power supply.
 
PC SPECS:
FX-8320
RX 580 ddr5 8GB
16 RAM
SSD Kingston 480gb
Redragon 500W 80 plus bronze (https://www.redragonzone.com/products/redragon-rgps-gc-ps001-500w-gaming-pc-power-supply)

GPU clock speed in games:
1250 MHz
VRAM speed:
2000 MHz
Temperature: 60° C

Crashes are random. I can play for hours before my games crash to desktop with the message error "Unreal Engine is exiting due to D3D device being lost" and AMD Adrenaline resets all video settings to default (also known as the Wattman crash error or driver timeout).

I tried different methods to fix the problem but none of them worked. Here's what I did:
  • I formatted my PC and did a clean install of my OS.
  • Reinstalled my GPU drivers.
  • Tried different versions of AMD Adrenaline. 20.11.2 / 22.3.1 / 22.11.2
  • Changed energy settings to High Performance in Power Management via the Control Panel. Turned off Fast Start Up, Sleep, Hibernate and Lock mode.
  • Disabled AMD Crash Defender via Device Manager and msconfig.
  • Ran the command sfc/scannow to scan the system and no files are corrupted.
  • Changed the TDR delay value to 10 or 60 decimal through regedit.
I don't know what else to do.

Some people said they had to undervolt their GPU to stop their games from crashing. I don't want to mess around with the GPU. It's still under warranty.
This may seem rather elementary, but did you run DDU before each driver installation? If not, I'd try that because installing new drivers over old drivers (even if you choose the "factory reset setting in the Adrenalin installer) causes all kinds of problems.

Previous posters referring to your PSU aren't wrong though. You shouldn't be using that piece of crap even if Polaris does only sip power.
 
  • Like
Reactions: th3reaperskullpt
Jan 21, 2023
5
1
10
This may seem rather elementary, but did you run DDU before each driver installation? If not, I'd try that because installing new drivers over old drivers (even if you choose the "factory reset setting in the Adrenalin installer) causes all kinds of problems.

Previous posters referring to your PSU aren't wrong though. You shouldn't be using that piece of crap even if Polaris does only sip power.
Yes, I ran DDU to remove everything and reinstaled the drivers.

I know you guys are trying to be helpful here but I really doubt the cause of the problem is low quality power supply. 500W is enough for a RX 580.

I'm going to return the GPU and they're gonna send me a new one. If I receive the same error again, then I'm gonna ask for a new power supply. I bought it last month and it's still under warranty too.
 
Last edited:
  • Like
Reactions: Avro Arrow
Yes, I ran DDU to remove everything and reinstall the drivers.

I know you guys are trying to be helpful here but I really doubt the cause of the problem is low quality power supply. 500W is enough for a RX 580.

I'm going to return the GPU and they're gonna send me a new one. If I receive the same error again, then I'm gonna ask for a new power supply. I bought it last month and it's still under warranty too.

You are welcome to be wrong. But don’t listen to the guy who’s fried ports on a motherboard using cheap power supplies. Don’t listen. If you come back saying you fried your system or your card, we told you so.
 
  • Like
Reactions: th3reaperskullpt

DSzymborski

Titan
Moderator
Yes, I ran DDU to remove everything and reinstall the drivers.

I know you guys are trying to be helpful here but I really doubt the cause of the problem is low quality power supply. 500W is enough for a RX 580.

I'm going to return the GPU and they're gonna send me a new one. If I receive the same error again, then I'm gonna ask for a new power supply. I bought it last month and it's still under warranty too.

The fact that you think getting a proper power supply is just choosing a specific wattage number is a pretty big issue. It's like shopping for used cars and only caring about the size of the gas tanks. It's a low-quality power supply: basic, cheap topology, cheap capacitors, etc.
 
  • Like
Reactions: th3reaperskullpt
Jan 21, 2023
5
1
10
The fact that you think getting a proper power supply is just choosing a specific wattage number is a pretty big issue. It's like shopping for used cars and only caring about the size of the gas tanks. It's a low-quality power supply: basic, cheap topology, cheap capacitors, etc.
Ok.

A low quality power supply, I get it.

Unfortunatelly, I don't have enough money to afford a better power supply. Not at the moment. So, I'm not gonna argue with that.

But what's that got to do with the error messages and games crashing? How exactly replacing my power supply for a high quality one will make the problem go away? Care to explain?

A lot of people are getting this error. It doesn't seem like it's hardware related. Not specifically. Even people with NVIDIA GPUs and powerful PCs are getting the same error.
 
Last edited:
It’s one of those things you want to start out with good clean power. If not having clean power can mess your system up. Have you ever gotten gas in your car that had some water in it? Had that happen once and the engine was sputtering etc.

Same kind of idea on a power supply. The cpu you have I think is a 125 watt cpu, the card will draw a decent amount as well. Also you figure that the better power supplies are 80% efficient, so a 500 watt can’t deliver 500 watts continuously. 80% of 500 is 400 watts. So if your pc is already close to the limit of 500, and then your power supply is a lower end unit, will probably be less efficient, you can see where I’m going. When your components try to start pulling the power they need and they can’t get enough, you can see where you may have issues.

Your power supply might be enough at idle when your system isn’t doing anything, but when your components start pulling power and you get voltage spikes, your power supply could be having issues and thus you see crashes etc. Fortunately it doesn’t sound like any of your components have been hurt as of yet.
 
  • Like
Reactions: th3reaperskullpt

th3reaperskullpt

Prominent
Dec 27, 2022
126
21
595
Ok.

A low quality power supply, I get it.

Unfortunatelly, I don't have enough money to afford a better power supply. Not at the moment. So, I'm not gonna argue with that.

But what's that got to do with the error messages and games crashing? How exactly replacing my power supply for a high quality one will make the problem go away? Care to explain?

A lot of people are getting this error. It doesn't seem like it's hardware related. Not specifically. Even people with NVIDIA GPUs and powerful PCs are getting the same error.
It is indisputable that the PSU is very bad in terms of quality and construction and may not be delivering enough energy or energy as promised to the components. It can cause exactly these power outages.
Most of the rest of the people who have the same problem have many different variables such as drivers or the windows itself corrupted, some misconfiguration at the BIOS level, some incompatibility with some component or just being derived from the version of the drive.
For example, a few weeks ago I opened a topic on the AMD forum regarding DX 12 because if I activate it my pc crashes the AMD application and among other problems, but it is related to the drive itself. But in your case we can put in case the fact that your PSU is not indicated.
But in your case it could be a problem derived from the PSU itself, because I doubt that other people have exactly the same components as you and the problems are different for each other.
 

th3reaperskullpt

Prominent
Dec 27, 2022
126
21
595
It’s one of those things you want to start out with good clean power. If not having clean power can mess your system up. Have you ever gotten gas in your car that had some water in it? Had that happen once and the engine was sputtering etc.

Same kind of idea on a power supply. The cpu you have I think is a 125 watt cpu, the card will draw a decent amount as well. Also you figure that the better power supplies are 80% efficient, so a 500 watt can’t deliver 500 watts continuously. 80% of 500 is 400 watts. So if your pc is already close to the limit of 500, and then your power supply is a lower end unit, will probably be less efficient, you can see where I’m going. When your components try to start pulling the power they need and they can’t get enough, you can see where you may have issues.

Your power supply might be enough at idle when your system isn’t doing anything, but when your components start pulling power and you get voltage spikes, your power supply could be having issues and thus you see crashes etc. Fortunately it doesn’t sound like any of your components have been hurt as of yet.

It looks like a white label power supply to me. I've been following the typical behavior of these power supplies, most of them can't even deliver 12v 230 watts without committing suicide along the way
 
Yes, I ran DDU to remove everything and reinstaled the drivers.

I know you guys are trying to be helpful here but I really doubt the cause of the problem is low quality power supply. 500W is enough for a RX 580.
It's not the amount of power that is being referred to but the quality of the power. Cheap PSUs can have poor sine waves and are prone to ripple effects because of cheap power regulators. It's rare, but these can contribute to a video card acting odd. Not enough power would mean that the card won't work at all. You may be right and it may be the card, we're just trying to help you see if it's something else because doing an RMA can be a pain in the butt.
I'm going to return the GPU and they're gonna send me a new one. If I receive the same error again, then I'm gonna ask for a new power supply. I bought it last month and it's still under warranty too.
Oh, well, if it's only a month old, then I'd say that you're probably right. Even cheap PSUs generally work properly out of the box. I was under the impression that you'd had it for a number of years. I think that your plan to RMA the card is most likely the best one in this case. That is, of course, unless they test the card, find nothing wrong with it and send it back to you but most companies don't bother with that.
 
  • Like
Reactions: th3reaperskullpt

Kona45primo

Honorable
Jan 16, 2021
523
143
9,890
IT's a long shot, but I know I've hear of ram conacts getting dirty and causing issues, wonder if that same could be true for GPU cards?

Might be worth a shot. It's probably a 11 year old rig give or take a few..

I still have my old FX-8120 system up and running. Upgraded to an SSD and gave it to my parents. Still kicking right along with the orginal Cooler Master Hyper 212 Evo cooler and Corsair GS700 power supply.