Is the ‘acoustid_fingerprint’ metadata required if the track already includes ‘acoustid_id’?

Since 'acoustid_‘fingerprint’ should never change once it’s included in the track, is there a reason MBP would need to use the website if I need to rename tracks or is ‘acoustid_id’ enough for MBP to find it?

Can I simply use ‘acoustid_id’, or should I use ‘acoustid_fingerprint’, or should I use both?

btw. I use ‘metaflac’ to clean my tags.


If you use Scan in Picard it does a search by AcoustID fingerprint on Usually it will run the fpcalc command on the file to calculate the fingerprint. But if the acoustid_fingerprint tag is present it uses the value directly. So it’s not really needed to store the acoustid_fingerprint tag, as the data can be easily calculated. Hence Picard by default does not store this tag.

But some users have a workflow where they e.g. calculate the fingerprint in advance by some other means and store it in the tag. Then they avoid the need to have Picard run fpcalc again.

acoustid_id is the ID returned as a result for the fingerprint lookup on It is not used for the lookup itself.


acoustid_id is merely informational - but it’s useful to me as it’s a confirmed acoust ID for my file¹ and thus for the particular recording. In case of merging recordings I have a verified acoust ID for comparison. There are usually many wrong acoust ID linked to a recording (unless there was a clean-up recently).

¹) The acoustid_id returned from a scan is not always the one your fingerprint submits to, but it’s certainly similar and its fingerprints can be compared to fingerprints of the recording to be merged.


Will MBP still send the track to if it has the ACOUSTID_ID tag, since it was already finger printed previously (like in a different session)?

I simply want to include which ever one is actually ‘required’ to only rescan tracks. (and why go to if MBP already has the acoustid_id.)
I don’t think I’m making sense… Can I just use MBP and ACOUSTID_ID to rescan my songs without MBP going to Maybe I should just forget tagging with the IDs)


Once you have the data in the track there is no need to re-Scan the track. The MBID will be enough to re-identify your track from the database if you just need to update data. And that is used in the Lookup.

Don’t press Scan and the AcoustIDs will stay untouched and no further lookups to Keeping the MBIDs in a track is enough for the links to stay to the MusicBrainz database and be found from a Lookup.

Don’t press “submit AcoustID” and Picard won’t send the data a second time.

Like @ernstlx I find the AcoustIDs useful when I want to work out where a track on a compilation was originally sourced from. I can visually reference that on screen, unlike a fingerprint.


If you use “Scan” this is always doing a lookup by fingerprint on the AcoustID server.

Theoretically it would also be possible to lookup by AcoustID, but again this would be a query to AcoustID API as the link between AcoustID and the MB recording IDs is stored on the AcoustID side.

Not sure what your use case is. But if you want to be able to retag your files without redoing all the matching again I recommend you have the MB recording and release MBID stored to the file tags and have “Ignore existing MBIDs” disabled. Both is the default behavior.

Not by default, but you can enable it in options, yes.