You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your issue present in the latest beta/pre-release?
This issue is present in the latest pre-release
Describe the Bug
The amd_usage function is used to change the encoder preset for AMF encoder. However, this seems to always be stuck on the lowest latency preset, no matter the option being chosen.
Expected Behavior
In 2024.911.215654 ( the latest pre-release I was using and where it worked, don't know when it broke), when the amd_usage option was changed to a slower preset (like low latency instead of ultra low latency), two expected behaviors were observed:
The encoding latency goes up because the chosen profile is slower
The GPU usage goes down substantially in encoding, with lower power consumption of the GPU
However, with this pre-release, this is no longer the case, latency doesn't go up and GPU usage doesn't go down, which may indicate that the used profile is not the correct one.
Additional Context
There is in the log a UseLowLatencyInternal: true and EncoderMFLowLatency: false when using the low latency preset which is the second fastest.
Is there an existing issue for this?
Is your issue described in the documentation?
Is your issue present in the latest beta/pre-release?
This issue is present in the latest pre-release
Describe the Bug
The amd_usage function is used to change the encoder preset for AMF encoder. However, this seems to always be stuck on the lowest latency preset, no matter the option being chosen.
Expected Behavior
In 2024.911.215654 ( the latest pre-release I was using and where it worked, don't know when it broke), when the amd_usage option was changed to a slower preset (like low latency instead of ultra low latency), two expected behaviors were observed:
However, with this pre-release, this is no longer the case, latency doesn't go up and GPU usage doesn't go down, which may indicate that the used profile is not the correct one.
Additional Context
There is in the log a UseLowLatencyInternal: true and EncoderMFLowLatency: false when using the low latency preset which is the second fastest.
Host Operating System
Windows
Operating System Version
Windows 11 24H2
Architecture
amd64/x86_64
Sunshine commit or version
v2025.212.40608
Package
Windows - installer (recommended)
GPU Type
AMD
GPU Model
RX 7900 XTX
GPU Driver/Mesa Version
24.2.1
Capture Method
Desktop Duplication API (Windows)
Config
Apps
Relevant log output
The text was updated successfully, but these errors were encountered: