+- Jellyfin Forum (https://forum.jellyfin.org)
+-- Forum: Support (https://forum.jellyfin.org/f-support)
+--- Forum: General Questions (https://forum.jellyfin.org/f-general-questions)
+--- Thread: I have a problem (/t-i-have-a-problem)
I have a problem - Homind - 2024-07-24
Hey!
I recently installed jellyfin on a Raspberry Pi 4 and absolutely love it!
I unfortunately have some files that contain multiple audio streams and multiple subtitle streams that I all need.
According to ffprobe all the audio streams are aac and all the subtitles are dvd_subtitle.
When I play the video, select the default audio stream and disable subs everything works great, since direct play is used.
But when I select any other audio track and/or a subtitle stream, the raspberry pi is fully occupied by trying to provide a transcoded stream for this single client.
Can I somehow prepare/extract the audio and subtitle streams so transcoding won't be necessary? I tried to extract one aac stream via ffmpeg and used it as an "external file". I was able to select it, but with external audio tracks the performance was significantly worse than just using the embedded ones and transcode.
Looking forward to suggestions and thanks for reading!
RE: I have a problem - 34626 - 2024-07-24
It's long time since i used a Raspberry Pi 4 for Jellyfin, due to the limitation you experience.
In jellyfin you can go to the dashboard -> Playback and change the hardware acceleration (I dont know if that will really improve anything).
But else, you might need to edit the config.txt on the raspberry pi and give the GPU 320M of ram + OC it - If you do this, it's on your own risk, the Pi will need cooling, i used the Argon One M2 case for mine (Just an example).
But i would recommend you to use another devices then a Raspberry Pi
RE: I have a problem - TheDreadPirate - 2024-07-24
Jellyfin/ffmpeg does not support HWA on RPi4's. Not that the RPi4 has any useful media engines anyway.
Many Jellyfin clients support direct playing the first audio and subtitle track (SRT/ASS subtitles only). If you select a secondary audio track it will require a remux on most clients, which your RPi should be able to handle. But if your subtitles are PGS, or some other image based subtitle, that would require burning in the sub. Which requires transcoding.
You either need hardware capable of handling transcoding or remove all the audio tracks you don't want so that there is only the one. And have SRT subtitles.