Semi-frequent driver timeout crashes
|
I've recently been struggling with driver timeout crashes. Many times a day, the game will lock up, screen goes black, and then I get booted to login with an error message popping up that AMD detected a driver timeout. I've tried fully wiping my GPU drivers and reinstalling them with no impact.
These crashes are annoying during gameplay, but they've completely crippled my ability to do uber lab. I'm trying to get my fourth ascendancy and so far I've lost the run 4 or 5 times in a row due to driver timeout. I've tried searching for other reports of this problem and most of the threads I have found are quite old. If anyone has any idea of what to do about this I'd appreciate it! IGNs Standard: Gyakufu | Gyakufuu Warbands: Tsukikage Tempest: Yamakage Last bumped on Jul 27, 2025, 1:42:50 PM
|
|
|
GPU driver crash, the first time this ever happened to me, and I have this system for 15 years.
It started pretty much at league start. Screen went black, no idea what was going on(I mean I suspected it from the get go), it comes back within 5-10 seconds and the game is not responding. So the only way is to close the PoE by whatever means necessary. The 3rd time this happened, NVIDIA app actually popped up with driver crash and sent info to support or whatever. So far it has crashed like a dozen or more times. I was using dx11 for a looong time since dx12 was memory hungry and Vulkan was leaky. I have tried Vulcan but loading into any, I repeat ANY instance takes 10 minutes so I dunno about that, but dx12 seems vastly more performant now. Can confirm this happens on dx11 and dx12. My GPU is GTX 750 ti the driver on the league start was 568 something, I don't really remember, so naturally I did a fresh install and now I have 576.80 and the issue is still present. There is an update for 577 but there is no mention about crashes, just some irrelevant new tech and game ready driver... So I don't see much point in bothering with another update since this issue is exclusive for PoE. After each crash, there is stuff in client log: A spam like this with presumably different ID and same hex(ID?), the hex number seems to be always same on every single log
Spoiler
2025/07/27 18:45:09 108589812 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null
2025/07/27 18:45:09 108589812 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:09 108589812 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:09 108589812 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:09 108589812 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:19 108600203 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:19 108600203 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:19 108600203 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/27 18:45:19 108600203 4ce63f2b [WARN Client 5648] [GPU TRAIL] Failed to create draw-call, LUT handle was null Here's an older log when I was using dx11:
Spoiler
2025/07/03 08:05:26 297458390 4ce63f2b [WARN Client 10476] [GPU TRAIL] Failed to create draw-call, LUT handle was null
2025/07/03 08:05:26 297458390 4ce63f2b [WARN Client 10476] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/03 08:05:26 297458390 4ce63f2b [WARN Client 10476] [GPU TRAIL] Failed to create draw-call, LUT handle was null 2025/07/03 08:05:32 297463828 edfb3208 [CRIT Client 10476] [D3D11] CreateShaderResourceViewEx: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464046 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464046 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464078 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464078 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464078 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464078 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464078 edfb3208 [CRIT Client 10476] [D3D11] CreateBuffer: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464093 edfb3208 [CRIT Client 10476] [D3D11] CreateComputeShader: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) 2025/07/03 08:05:32 297464093 95b054a0 [CRIT Client 10476] [SHADER] Failed to load shader (error: CreateComputeShader: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.)) ### there's a spam of this so I'll pick some different ones 2025/07/03 08:05:38 297470046 95b054a0 [CRIT Client 10476] [SHADER] Failed to load shader (error: CreateComputeShader: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.)) [update_base_lifetime_pps_seed_spawn_load_colour_scale_flip_rotate_rotate_gravity_velocity_cylinder_rotate_random_update_lock_mo] 2025/07/03 08:05:38 297470156 f2498ec3 [CRIT Client 10476] [EXCEPTION] CreateShaderResourceViewEx: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action. (GetDeviceRemovedReason: The GPU will not respond to more commands, most likely because of an invalid command passed by the calling application.) Seen people talk about event viewer so here's an Error that seems to be related: Error, source nvlddmkm
Spoiler
The description for Event ID 153 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
Anyway, bottom line is, last league was fine, never had a GPU driver crash with my 15 year old PC. KranglingKearKames
https://i.imgur.com/zB2RgAs.mp4 https://imgur.com/a/mafhrmH |
|
