Tagging disc ID's

[evasv]

There are a bunch of (disc) ID's that some people tag into their flac files:


ACCURATERIPID
ACCURATERIPDISCID

MUSICBRAINZ_ARTISTID
MUSICBRAINZ_ALBUMID
MUSICBRAINZ_TRACKID

ACOUSTID_ID


Just to mention a few.

Do you guys tag the ID's above, or other disc/track ID's? What is the purpose?

Do online services respond better with these ID's?

Or are they tagged just in case you will ever need them in the future?

[evasv]

Maybe I should explain a little bit more. I use a lot of tags, both standard and custom, aiming to be able to browse and seek in my collection in various ways. But I am not the "librarian type" who wants to categorize every possible thing just for the sake of it.

For example I use tags as Country, Producer and Label as they actually help me to find music. But I don't (at least for now) use Catalognumber, Barcode or Musicbrainz_* tags as they are primarily for identification and pointless to have for browsing and searching.

But I'm debating with myself if I should add those ID tags anyway when I re-tag my collection. Could there be some reason to have them that I have missed? For online services, if I ever need them again after I've tagged my collection - do they give a better service if you have ID tags? I am somewhat concerned that if I import those ID's they could be inaccurate and actually make things worse.

And besides online services, could there be other benefits with ID tags in your music files that I have missed?

Accurateripid for example - if the files are already accurately ripped, is there any purpose other than academic to have that ID tagged? Does anything outside accuraterip / dbpoweramp use that tag?

I'd be very thankful if someone could share his or her thoughts on this (or maybe just point out an ID that seems to be more relevant to have than others).

[garym]

accurateripID and discid are created by dbpa (and maybe other rippers) that use ACCURATERIP to confirm bitperfect rips. (EDIT: and you're correct....I don't see much advantage of the tag after the fact; although PerfectTunes can check your already ripped files back to the AR database and one option is to use the tag info rather than rechecking the audio back to AR.

the Musicbrainz tags are added by the tagger "musicbrainz" (not added by dbpa). I have the Accuraterip tags (because added by dbpa). But I don't use the musicbrainz tags. not sure what Acoustid_id is.
https://picard.musicbrainz.org/

EDIT: ACOUSTID seems to be a music fingerprinting program output. See
https://acoustid.org/

[evasv]

Thanks for your input. I guess these ID's aren't widely used, but many have them anyway because they are automatically tagged by some program.

Myself I wiped my collection from tags years ago, so if I want the ID's now I'd have to tag them manually or import them from online databases.

Importing seems risky - how can you trust that you get the right ID's when you search in the databases only by album & album artist? And even if you get a "perfect match", there's still the risk that the database ID is wrong.

I would rather calculate the ID's myself if I knew how. The best thing would be to have program(s) calculating and tagging the ID's without contacting the databases. A quick test I did indicated that Picard seems to calculate the acoustid but imports the musicbrainz ID's.

Perfecttunes could calculate & tag the accuraterip ID, but this ID is probably of no practical use after ripping. If you want to recheck the integrity of your already ripped files back to the database you'd want to recheck the binary data too I think. But it would be neat though if other databases started to use the accuraterip ID instead of creating their own, since the accuraterip ID must surely be the most widespread one.

In my wildest dream I'd like an ID that could help getting the right lyrics, biographies, art etc from online databases/services when you're playing your music. But it would require a unified ID to work properly.

[garym]

As far as I know, there is no way for you to calculate these IDs yourself. You don't have them now, so my advice is to forget about them and just make sure you have well tagged music with regard to artist, album, title, etc. If you are a fanatic about metadata, you might consider using ROON. The Roon Labs server has tremendous metadata/info that it matches automatically (and it does very well) with your library. A very "rich" set of metadata. But it's not free.
https://roonlabs.com/

[Dat Ei]

Perfecttunes could calculate & tag the accuraterip ID, but this ID is probably of no practical use after ripping.

Could it really calculate that accuraterip ID? I don't think so. Checksums are unique - most of the times. And what if this rip is corrupted and you get a different checksum? The metadata can be ok, but don't have to. So the accuraterip ID is a unique key representing just one special CD incl. its tracks and checksums.


Dat Ei

[garym]

Could it really calculate that accuraterip ID? I don't think so. Checksums are unique - most of the times. And what if this rip is corrupted and you get a different checksum? The metadata can be ok, but don't have to. So the accuraterip ID is a unique key representing just one special CD incl. its tracks and checksums.


Dat Ei

Yeah, I was thinking the same thing. PerfectTunes can *check* your rips and tell you if they are AR matching, but not add a tag value. But as the OP said, why do you really need the tag value if you know the rip matches in AR?

[evasv]

If you are a fanatic about metadata

At least I'm more fanatic about accurate metadata than large amounts of metadata. So maybe I'm better off not striving for too much and forget about these ID's as you say.

[evasv]

Could it really calculate that accuraterip ID? I don't think so.

Sorry, my mistake. I was assuming it could.

And you are right about the unique key. I would want matches to be 100%, if they're not there's no point.

[garym]

At least I'm more fanatic about accurate metadata than large amounts of metadata. So maybe I'm better off not striving for too much and forget about these ID's as you say.

agree