Getting the details right

Most things identify themselves. This guide is about the ones that do not, and about taking manual control when you want it.

Everything here is done from a web browser and is owner-only. The TV and phone apps show the results; they do not edit the library.

Fixing a wrong match

Open the item, then ⋯ → Identify.

You get one dialog that can do three different jobs:

• Search and pick. Type a better title, optionally a year, and choose from the results. Searching across films and TV both, so you are not stuck inside the wrong category.
• Change what kind of thing it is. A film that is really an episode of a show can be reclassified into that show, in the same dialog. This is the one that usually cannot be fixed anywhere else, and it is why the dialog exists.
• Enter it by hand. For home videos, a concert recording, or anything that is genuinely not in any database, type the title, year, and description yourself.

Once you have identified something, it stays identified. Later scans do not undo it.

Editing fields by hand

⋯ → Edit gives you every field: title, sort title, year, description, tagline, genres, studio, content rating.

Each field you change is locked individually. When the server later refreshes that title, it updates everything else and leaves your edits alone. So you can correct one bad description without freezing the rest of the record in time.

There is an unlock control per field for when you change your mind and want the automatic value back.

Refreshing

⋯ → Refresh pulls the record again, useful when a title was matched before the database had good artwork for it, or when a show’s episode list has filled in since.

New episodes of a currently-airing show are a special case, and one you do not have to do anything about. When an episode airs before the metadata database has indexed it, the server marks it and quietly tries again on its own for the next few weeks. It fills itself in.

Artwork

Posters, backdrops, and episode stills arrive automatically and are cached on your server, so they load instantly and keep working when the internet does not.

If you imported from Plex, any posters you had uploaded there came across with everything else.

Several cuts of one film

Two files that are the same title get merged into one entry with a picker, rather than sitting next to each other as duplicates.

The clean way is to tag the filenames:

Blade Runner (1982) {edition-Theatrical}.mkv
Blade Runner (1982) {edition-Final Cut}.mkv

If you cannot rename the files, open the title and use ⋯ → Editions to name each attached file yourself.

How it then behaves:

• Pressing Play offers the choice, with each cut’s runtime and rating: Theatrical · R · 3h 22m / Final Cut · Unrated · 3h 47m
• Your resume point remembers which cut it belongs to. Keep Watching takes you back into that one.
• Picking the other cut starts it fresh, and from then on that is the one you are resuming.
• Watching either one credits the title as watched.
• One edition is the default, used anywhere there is no picker: auto-play of a next episode, downloads, and older app versions. Set it in the editions editor. It is seeded to the one tagged Theatrical, or the first one found.

This works per episode too, which is what you want for a show that shipped a colour version and a black-and-white version of the same episodes. Pick once, and the choice carries as episodes auto-advance.

Cast and crew

Every title carries its full cast and crew. Tapping an actor opens their page, which lists everything of theirs that is in your library, films and shows together. Directors and writers get the same treatment.

This is also what makes rules like “comedies starring Matt Damon” possible in smart collections.

Age ratings and which country’s system you see

Titles carry certifications from every country that rated them, not just one. Settings → Libraries → Ratings country chooses which system gets displayed, so a UK household sees 12A and 15 and a US household sees PG-13 and R for exactly the same files.

We draw our own rating chips rather than using the official artwork, which is registered trademark material we have no licence to reproduce.

The same certifications are what parental controls compare against, and having all countries means a film certified only in Germany still gates correctly in a US home.

Files that are hard work to play

Some files fight every device in the house: AV1 and VP9 web rips, 10-bit HEVC on a server with no hardware decoder, old VC-1 and MPEG-2, or audio tracks nothing can play directly. Every time someone watches one, the server has to re-encode it live, which is slow, hot, and sometimes stutters.

Optimize Media pre-encodes those files once into a friendly copy, at the same resolution and the same number of audio channels, so every later play is cheap.

• The server profiles its own hardware and scores your files. Titles that would benefit show a warning with a plain-English reason and an Optimize button.
• Settings → Optimize Media lists every flagged file so you can select a batch and queue it.
• Jobs run only when nobody is streaming. If someone starts watching mid-job, the job slows down rather than being thrown away.
• Your originals are never deleted. The optimized copy sits in an Optimized folder next to the source, and playback picks whichever copy is cheapest for the device asking.