Question about using MB on large number of files by processing groups of songs at a time


#1

I am a total novice at using Musicbrainz and don’t fully understand how it works. I have a folder with a large number of music files that need tagging and sorting into folders. I can select a group of files at a time for it to process so that MB doesn’t choke on the whole folder of files. I assume that if I set MB to move files it will arrange them in folders by artist, album, etc. What I don’t understand is how it will work on subsequent groups of files that I give it to process. Will it know about previous folder groupings and use them to sort additional files? If there are instructions somewhere for how to process large sets of music files in batches I’d appreciate a pointer. Thanks.


#2

First try and work with small batches of around 100 files or less.
It is a good idea to create a backup before you start just in case you don’t like the results.

Picard is designed to work with the musicbrainz database so a release needs to exist in the database for it to find it.
Disable all plugins first run, your aim should be trying to match the recording to an item in the database.
Once you have saved the file it will contain musicbrainz identifiers making it able to find the information again so later you can enable all the plugins you want to add more information.

A recording can be on multiple releases so you can find random compilations as well as the album or single by that artist so you should try and get as close as you can when tagging music.

Add a small number of files each run.
Click Cluster, this tries to group files by tags and should group songs from the same cd together.
Click the lookup button when selecting the cluster. This uses tags to match files in the database and should come up with a reasonable guess for the release.
Before saving make sure you have the right release, right click on the release on the right side and try and find the version you have.

For songs that don’t have good enough tags to start with there is the scan button.
This looks up items based on thair fingerprint.
As recordings can be common to multiple releases you need to work out if you have the right album before you save it.


#3

The specific file and folder naming structure will be consistent between runs of Picard. So yes, if you pick 50 tracks tag and move them. And then the next 50 tracks overlap some of those albums, then Picard will move everything into that same pattern.

Have a look into the settings and you’ll fine the File Naming section. This will allow you to control the folder and track naming so it fits the exact pattern you like.

As @dns_server mentions, there are various MBIDs that are being added to the tags so the next time you look at your sorted folders they will be quickly identified by Picard and make further tagging and sorting smoother.

These IDs are also handy as a few days into doing this cleanup you’ll all most certainly change your mind about how you want some of the details. As you learn more about your music files and how MB works, no doubt you’ll then see better ways of naming your files.

https://picard.musicbrainz.org/docs/options/#file_naming