Hardware acceleration on Synology DS224+

[The Bow-Tie Guy]

Thanks very much, but what didn´t work is the implementation with docker container to tell jellyfin what to do and to give it access to the gpu.... that´s what I was asking for in my first post.
When I now only activate quick sync and go with what the GeminiLake column says some videos won´t start.

So do I have to add this at any point to docker container?
services:
jellyfin:
image: jellyfin/jellyfin:10.8.11
container_name: jellyfin
# user: uid:gid
network_mode: 'host'
volumes:
- ./configconfig
- ./cachecache
- /volume1/Media/Filmemovies:ro
- /volume1/Media/TV Showstv_shows:ro
devices:
- /dev/dri/renderD128dev/dri/renderD128
- /dev/dri/card0dev/dri/card0
#note: uncomment these lines in devices to allow for HWA to work on Synology units with an iGPU
restart: 'unless-stopped'

Cause that was what didn´t work and it was frustrating cause I watched so many tutorials for that. Other question .... do I have to re-install everything to make work? Open a project in the docker container and then tell jellyfin what to do and give it access to the gpu. Oh when I tried I wasn´t able to open port 8096, what normally works automatically when you install jellyfin with adding those lines I posted above.

Thanks for the help!

[The Bow-Tie Guy]

So I tried that now, without adding the command lines from above (couldn´t get it to work), well and some mkv container won't play.

[TheDreadPirate]

What version of jellyfin are you using. Because the docker compose you're sharing says 10.8.11, but the ffmpeg log you shared is using ffmpeg 7, which is only available starting with 10.10.

So I'm not sure you're looking at the right configuration page for the instance of Jellyfin you are actually running.

[The Bow-Tie Guy]

The version I showed in the command was an old version, I got from a tutorial. I use now 10.10.7 but without adding the command line adding and allowing the container to access the gpu.

[The Bow-Tie Guy]

So I redid everything now, installed the project in the docker container with the command lines and the latest jellyfin... and it seems to work. The cpu is between 40 and 50 now instead before it was 90.

Can I post some pictures later today that you take a look at it?