Monster Hunter Wilds

Monster Hunter Wilds

View Stats:
Fatal D3D error (25, DXGI_ERROR_DEVICE_HUNG, 0x887a0006)
anyone has a fix for this was not able to play the beta because of it, now i tried to run the benchmark and im having the same problem game just crashes.
all drivers up to date, new windows install since 2-3 month ago.
GPU 4090
CPU 7800 x3d
< >
Showing 1-15 of 60 comments
This will happen when a GPU call is made to the API and the driver times out. Use DDU and/or roll back your driver

I can almost guarantee you if you look in event viewer you will see an nvlddmkm error

Also turn off Nvidia Share (Formerly known as Shadowplay) - this hooks into Nvidia's telemetry container during API runtime.
Last edited by RakataEpsilon; Feb 14 @ 1:51pm
kohlrak Feb 15 @ 6:02am 
Originally posted by Genjur0:
anyone has a fix for this was not able to play the beta because of it, now i tried to run the benchmark and im having the same problem game just crashes.
all drivers up to date, new windows install since 2-3 month ago.
GPU 4090
CPU 7800 x3d

Same issue on AMD (but linux equivalent). There seems to be a shader acting screwy.
Foxrun Feb 15 @ 7:59am 
Originally posted by Genjur0:
anyone has a fix for this was not able to play the beta because of it, now i tried to run the benchmark and im having the same problem game just crashes.
all drivers up to date, new windows install since 2-3 month ago.
GPU 4090
CPU 7800 x3d
Get rid of any overclock on your GPU.
Originally posted by kohlrak:
Originally posted by Genjur0:
anyone has a fix for this was not able to play the beta because of it, now i tried to run the benchmark and im having the same problem game just crashes.
all drivers up to date, new windows install since 2-3 month ago.
GPU 4090
CPU 7800 x3d

Same issue on AMD (but linux equivalent). There seems to be a shader acting screwy.
You're a on linux...


Can you not dual boot?
kohlrak Feb 15 @ 8:24am 
Originally posted by GamingWithSilvertail:
Originally posted by kohlrak:

Same issue on AMD (but linux equivalent). There seems to be a shader acting screwy.
You're a on linux...


Can you not dual boot?

Don't have windows with which to dual boot. Kinda irrelevant, too, the only suggested tool is for nvidia, not AMD.
Genjur0 Feb 15 @ 4:50pm 
Originally posted by RakataEpsilon:
This will happen when a GPU call is made to the API and the driver times out. Use DDU and/or roll back your driver

I can almost guarantee you if you look in event viewer you will see an nvlddmkm error

Also turn off Nvidia Share (Formerly known as Shadowplay) - this hooks into Nvidia's telemetry container during API runtime.
thank you, went back to one of the older drivers and it work.
Guess ♥♥♥♥ nvidia drivers again, starting to turn into old AMD.


Originally posted by Foxrun:
Originally posted by Genjur0:
anyone has a fix for this was not able to play the beta because of it, now i tried to run the benchmark and im having the same problem game just crashes.
all drivers up to date, new windows install since 2-3 month ago.
GPU 4090
CPU 7800 x3d
Get rid of any overclock on your GPU.
no OC was done, all stock
Originally posted by kohlrak:
Originally posted by GamingWithSilvertail:
You're a on linux...


Can you not dual boot?

Don't have windows with which to dual boot. Kinda irrelevant, too, the only suggested tool is for nvidia, not AMD.
Tis not irrelevant. CC doenst provide natural support for linux, and it could be many thinks. Is actually perfect a debug if your on linux.
Last edited by GamingWithSilvertail; Feb 15 @ 5:16pm
Genjur0 Feb 15 @ 5:31pm 
now after 1h of playtime crashed with this error
Fatal D3D error (24, DXGI_ERROR_DEVICE_REMOVED, 0x887a0005)
I believe these Nvidia issues were being caused by Hardware Acceleration, needs to be turned off in NV Control Panel/Windows etc.
kohlrak Feb 15 @ 9:38pm 
Originally posted by Goblin:
I believe these Nvidia issues were being caused by Hardware Acceleration, needs to be turned off in NV Control Panel/Windows etc.
Without hard work acceleration your effectively turning your GPU off. You're going to have to be more specific than that
AmaiAmai Feb 15 @ 10:39pm 
Tell the game's programmers about it, I am sure they are going to find and fix the routine causing it within a week and compile a new version of the game in that same time period (not).

That error is because they are incompetent at using low level hardware accelerated routines that are effectively running over and over with no result The operating system sees that no work is being done (no changes to values due to a buggy routine), but the device is "busy" and eventually it terminates the program after it reaches the timeout threshold and says the GPU is hung.

What you can do is adjust this timing if you find it too strict, you can ask Nvidia what registry entries they used to set this value on the driver side and adjust it, or find the routine yourself.

Now, with modern games this happens often because of raytracing routines that are broken and/or buggy, the other are hardware-accelerated reflections you can simply try disabling both of those and seeing if it runs.

But again, the problem is a problem for the developer to fix.
kohlrak Feb 16 @ 1:13am 
Originally posted by AmaiAmai:
Tell the game's programmers about it, I am sure they are going to find and fix the routine causing it within a week and compile a new version of the game in that same time period (not).

That error is because they are incompetent at using low level hardware accelerated routines that are effectively running over and over with no result The operating system sees that no work is being done (no changes to values due to a buggy routine), but the device is "busy" and eventually it terminates the program after it reaches the timeout threshold and says the GPU is hung.

What you can do is adjust this timing if you find it too strict, you can ask Nvidia what registry entries they used to set this value on the driver side and adjust it, or find the routine yourself.

Now, with modern games this happens often because of raytracing routines that are broken and/or buggy, the other are hardware-accelerated reflections you can simply try disabling both of those and seeing if it runs.

But again, the problem is a problem for the developer to fix.

I'm impressed and I'm going to have to try your solution when I get home I'm assuming that screen space reflections water effect s I don't think Ray tracing is an issue in my case cuz I disabled FSR to see if that would improve things since I'm rendering it native resolution anyway.... Disable variable rate shaders of course because that can enable things that are otherwise disabled and vice versa... Anything else I'm missing?
Originally posted by AmaiAmai:
Tell the game's programmers about it, I am sure they are going to find and fix the routine causing it within a week and compile a new version of the game in that same time period (not).

That error is because they are incompetent at using low level hardware accelerated routines that are effectively running over and over with no result The operating system sees that no work is being done (no changes to values due to a buggy routine), but the device is "busy" and eventually it terminates the program after it reaches the timeout threshold and says the GPU is hung.

What you can do is adjust this timing if you find it too strict, you can ask Nvidia what registry entries they used to set this value on the driver side and adjust it, or find the routine yourself.

Now, with modern games this happens often because of raytracing routines that are broken and/or buggy, the other are hardware-accelerated reflections you can simply try disabling both of those and seeing if it runs.

But again, the problem is a problem for the developer to fix.

Well...it works on Radeon GPU's.

So, is it a game dev issue, or has Nvidia just borked yet another driver release?
Originally posted by AmaiAmai:
Tell the game's programmers about it, I am sure they are going to find and fix the routine causing it within a week and compile a new version of the game in that same time period (not).

That error is because they are incompetent at using low level hardware accelerated routines that are effectively running over and over with no result The operating system sees that no work is being done (no changes to values due to a buggy routine), but the device is "busy" and eventually it terminates the program after it reaches the timeout threshold and says the GPU is hung.

What you can do is adjust this timing if you find it too strict, you can ask Nvidia what registry entries they used to set this value on the driver side and adjust it, or find the routine yourself.

Now, with modern games this happens often because of raytracing routines that are broken and/or buggy, the other are hardware-accelerated reflections you can simply try disabling both of those and seeing if it runs.

But again, the problem is a problem for the developer to fix.
There's litteralkly modified windows versions goin around and you blame the dev if the game crashes cause people use modified windows? :p

I dont think thats a developers fault. That's a user end fault. Thats why I tell people on linux users to dual boot on windows so they can actually see if something is a problem or not, since the game natively is not supported on linux, you rely on Proton, therefore thats a valve issue.
Last edited by GamingWithSilvertail; Feb 16 @ 2:38am
kohlrak Feb 16 @ 4:08am 
Originally posted by Goblin:
Originally posted by AmaiAmai:
Tell the game's programmers about it, I am sure they are going to find and fix the routine causing it within a week and compile a new version of the game in that same time period (not).

That error is because they are incompetent at using low level hardware accelerated routines that are effectively running over and over with no result The operating system sees that no work is being done (no changes to values due to a buggy routine), but the device is "busy" and eventually it terminates the program after it reaches the timeout threshold and says the GPU is hung.

What you can do is adjust this timing if you find it too strict, you can ask Nvidia what registry entries they used to set this value on the driver side and adjust it, or find the routine yourself.

Now, with modern games this happens often because of raytracing routines that are broken and/or buggy, the other are hardware-accelerated reflections you can simply try disabling both of those and seeing if it runs.

But again, the problem is a problem for the developer to fix.

Well...it works on Radeon GPU's.

So, is it a game dev issue, or has Nvidia just borked yet another driver release?
No I have the Linux equivalent on my AMD GPU. It's definitely a shader and I knew that before this post but to know that this guy seems to have a solution that might work I think I'm going to go ahead and try it and report back my results.
< >
Showing 1-15 of 60 comments
Per page: 1530 50