I rip all movies with the best possible audio track since I have a home cinema system that can handle it.
I do however also include the Dolby Digital and DTS lossy tracks for use when streaming away from home.
I also share my library with family members and I have several times asked, or told them, to change the audio track to the DTS or DD since they do not have a system that handles lossless audio well. Hence it is not necessary.
The problem is that this information does not sink in, and to be honest I can’t blame them. My father is over 70. He just does not know what I am talking about. The same goes for most other users of my server. They are used to Netflix and Disney+ and do not think about audio tracks.
Usually this is not a real problem, I have a good upload speed on my internet, and the server seems to handle the necessary transcoding well.
Still, I was wondering if I could set the DD or DTS tracks as default. This way I can change to the lossless track when needed instead of asking all others to change to a lossy format.
I really hope this is possible somehow. I feel it should be possible at least. Maybe a future update?
Almost everything with Plex is controlled by the player.
The player is subject to the playback device (TV, tablet, phone) limitations.
You, as the server owner/admin can limit the bandwidth because you know and have control over that.
You don’t have control over the devices playing from you.
Therefore, the player apps look at the device they’re connected to / resident on, then look at the app’s Quality settings.
When the app makes a playback request,
The app tells PMS what it wants to play
The app tells PMS what it’s capable of (Audio/Video/subtitles)
The two negotiate what to play to achieve best experience
That’s what comes out of PMS.
Example:
You have a full BluRay UHD rip at 120 Mbps 4K but
– the TV is only 1080p (no UHD… it’s SDR only)
You have Dolby Atmos audio
– the TV only has its two speakers.
PMS looks at what’s available to send (different versions of the file)
– sees there is only the 4K, Dolby Atmos file
– Starts a transcode session to
– * Convert the 4K to 1080 & tone map the UHD to SDR
– * Convert the Dolby Atmos to Opus stereo (or whichever audio codec the TV wants)
If you had imposed a bandwidth limitation, PMS and app would iterate one time at step 3 to satisify the bandwidth limitation AND select something the TV can play.
Adapt the way how you mux your files, so that the lossy sound track always comes before the high-bandwidth audio track.
I am creating always a stereo AAC track which I am sorting before the original audio.
MKVtoolnix makes this very easy – both the sorting of tracks, and inserting additional tracks without the need to re-encode all other tracks.
With that strategy, Plex clients will by default pick the first audio track (with the right language tag, if automatic track selection is activated). And in my case that is the most compatible and bandwidth-saving AAC stereo track.
The user needs to explicitly select a different track, if (s)he wants to use surround.
This manual selection is then stored on the server, only for that user.
If I understand you correctly the server will then need to have several different versions of the movie. It does not care that a compatible audio track is included inside the specific mkv file? That is somewhat a waste of hard disk space I would say?
In your example above the server will then start to convert the Dolby Atmos to Opus instead of using the Dolby Digital track that is already included in the mkv file. The problem is, if I understand it correctly, that the server only looked for another version of the movie and did not find this. Even though the file it tries to play indeed would be compatible in itself if you see what I mean?
Edit: Also, my current settings say Internet upload speed 245 Mbps (I have 250 up and down) and Limit remote stream bitrate Original (No limit).
All the videos I share are at 1080p maximum so 4K is not necessary. If I change this to 20 Mbps (1080p) that will force a transcode if the video, including audio, has a bandwith higher than 20 Mbps?
An example is the movie Hacksaw Ridge that runs ar 39 Mbps with 1080p and DTS HD MA 5.1 track.
This will certainly be a solution if it works that way!
Now I only have one problem. All my movies are already muxed with the lossless audio track first (since that is how it is done when ripping the movie I would assume?)
Is there an easy way to fix this without having to remove the movie from the Plex library folder, reorder and remux, and then finally moving it back into the Plex folder? Then I would need to adjust those settings that where changed like poster, sort order and so on…
I fear this fix could take some time to complete…
If you keep the file name and folder the same and disable library scans during the copy process, everything should remain the same. There might be a new detection of credits during maintenance though.
I just tried the following with what appears to be success
1 - Turned off “Scan my library automatically”
2 - Adding two test movies to MKVToolNix. I rearranged so that the DTS and AC-3 tracks were above their lossless counterpart. The output files where saved on another disk.
3 - When both files where complete I moved them into the Plex library accepting to overwrite since they both had the same name as those files already in the library.
4 - I did a test, but the lossless audio was still selected.
5 - I then Scanned the library files
6 - The two movies now appear with the DTS and AC-3 audio as selected
7 - All collections, posters, naming and sort order appear to be intact!
Technically, you need to trigger “Analyze” after swapping out the files.
remark: you might want to add a title to one of the English PGS subtitle tracks, to distinguish them. This could be “SDH” for a CC-type of sub, or setting the “forced” flag if applicable.
I tried to Analyze, first the entire library, then the two movies. I can’t see that I see any change, or anything happens at all. Is this a scheduled task completed later?
Yes, I’ve considered differentiate the English subtitles. However, I normally do not use them as long as there is a Norwegian subtitle available. They are there more “just in case”. If there are more than two Norwegian subtitles available I do name them.
While on that topic - is it possible to easily identify the SDH or commentary subtitle among the PGS files here in MKVToolNix?
Unfortunately not.
Sometimes the number of lines is an indicator.
You can see these with tools like MediaInfo
On a movie, if you have two tracks with several 100’s of lines each and only a small difference, you might have a regular and an SDH track on your hands.
If you have one that is much much smaller, it is probably a forced track.
But this is still just guess work. A “commentary” track can sometimes appear as big or even bigger than an SDH track.
For a safe determination of the subtitle type, you need to actually “read” the subs.
I use Subtitle Edit for that.
I’ve used an app called FileFlows to re-organise (and reencode/downmix if necessary, or a simple 2.0 or 5.1 track is missing) audio tracks to make dumber players auto-select a compatible feed.
once its set up it can just sit there and automatically fix files as needed.
