Jellyfin Forum
Tone Map Dolby Vision to SDR using 100-nit L2 Trim - Printable Version

+- 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: Tone Map Dolby Vision to SDR using 100-nit L2 Trim (/t-tone-map-dolby-vision-to-sdr-using-100-nit-l2-trim)



Tone Map Dolby Vision to SDR using 100-nit L2 Trim - Speedy - 2023-12-23

Does Jellyfin use Dolby Vision's 100-nit L2 Trim when tone mapping Dolby Vision media to Rec. 709 SDR?
It seems like the L1 metadata is being used, but not whether the L2 or L8 trims are being used.

Is this limited to specific color spaces? e.g., does it do it for both the IPTPQc2 (Profile 5) and YCbCr (Profile 7 & 8 color spaces?

I've run across the following but neither includes this information:
Thank you!


RE: Tone Map Dolby Vision to SDR using 100-nit L2 Trim - nyanmisaka - 2023-12-24

It is not. Currently you can get the free Dolby Vision playback support in open source players/renderers (such as mpv/libplacebo/mpc-be/jellyfin) is limited to the reshaping part of the dynamic metadata, while other parts such as L2 Trim are not yet supported. This already ensures that users will not be troubled by wrong colors when playing Profile 5 videos on SDR screens. You have to understand that there are limits to reverse engineering from the community, and if you want a premium Dolby Vision support buy a certified player and avoid transcoding.

https://code.videolan.org/videolan/libplacebo/-/merge_requests/207


RE: Tone Map Dolby Vision to SDR using 100-nit L2 Trim - Speedy - 2023-12-24

(2023-12-24, 01:21 PM)nyanmisaka Wrote: Currently you can get the free Dolby Vision playback support in open source players/renderers (such as mpv/libplacebo/mpc-be/jellyfin) is limited to the reshaping part of the dynamic metadata
@nyanmisaka thanks for the reply!

Just to confirm in technical language... when you say "limited to the reshaping part of the dynamic metadata", is Jelyfine (and other open source projects like mpv) currently only utilizing the L1 metadata and determining/interpolating tone mapping based on that (this seems very similar to how HDR10+ metadata would typically be used. 

Quote:L1 – Level 1 – (min, avg, max):
The Dolby Vision analysis produces three values per shot (or per frame) to describe the image and its dynamic range.

Thanks!


RE: Tone Map Dolby Vision to SDR using 100-nit L2 Trim - Speedy - 2024-01-22

Bumping my previous message/question