New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Support]: ffmpeg still decoding stream with detection disabled #8504
Comments
Even when detect is disabled the stream is still decoded, if you don't provide a sub stream then the main stream will be decoded which will increase CPU usage. |
@NickM-27 Is this expected behavior or a known issue? Why does Frigate need to decode the stream if I’m not using it for detection purposes? If it’s expected then perhaps we add a note to the docs that states the stream will be decoded regardless and this can significantly impact CPU usage? |
The decoded stream is used for other things like:
There is a feature request to make this not always be the case but as outlined above many things depend on this decoded stream so it will be complicated to implement. The docs can be updated |
Describe the problem you are having
I have two same model cameras (Wyze Cam v3, being broadcast through docker-wyze-bridge)- one was setup with detection and a low-res substream (for detection purposes), while the other is setup without detection (just recording). Both cameras are outputting the same resolution. I noticed the one without detection had significantly higher average CPU usage 50%+ while the other was consistently below 10%. This feels weird to be expected behavior because if I enable detection and setup the low-res substream on the other camera, its CPU usage drops to that of the other camera.
Version
0.12.1-367D724
Frigate config file
Relevant log output
FFprobe output from your camera
Frigate stats
Operating system
Debian
Install method
Docker Compose
Coral version
USB
Network connection
Wired
Camera make and model
Wyze Cam V3 via docker wyze bridge
Any other information that may be helpful
No response
The text was updated successfully, but these errors were encountered: