I’,m a newbie and relieved to find something to start getting me organised!
I want to know what to do with the unclustered tracks.
Ive loaded a folder with all my mp3s dumped in it into the the left hand pane.
I clustered/looked up/scanned and now i have 676 unclustered tracks and 179 clusters.
I highlighted and save everything in the right hand pane, I have rename/move/save tags all highlighted and pointed to a new directory called ‘organised music with folders’ and a populated folder structure has appeared there.
(I actually saved too early - i notice ‘pending files’ is still counting down but i dont know what part of the process is still ongoing)
I’ll wait until the activity is finished but what do i do next? I assume all the clusters have been saved in folders under their artists and titles but what about the unclustered files (I want to call them orphans). Are they collected somewhere? Do I need to save them? Is there any further processing i can do on them?
Firstly, be a little bit careful! You can’t undo changes to tags, and I wouldn’t run too many files through at once without double checking their matches in the right-hand column.
How Cluster works: It groups your files into albums, based on the existing ‘album’ tag, or sometimes the folder they’re in if that’s lacking and Picard thinks it makes sense
If you have a bunch of unclustered files you can still hit lookup (or scan as a last resort) on all of them and then hit save when they’re matched. But you’re going to end up with a lot of album folders with just one track in them as your files are matched to different albums and compilations. You can change the save behavior to not generate album folders at this stage of course, but you will still end up with a lot of single song albums in your player. This might be correct if you do have a bunch of incomplete albums/odd tracks in your collection.
If you think these files do actually belong to full albums then I would start working album by album, and use drag and drop to put them all into the same clusters or into the same albu. on the right hand pane, before hitting save.
that’s brilliant. This is only day 1 so a lot to learn. Yesterday i had nearly 60k files due to carelessness and ill discipline on my part! I’ve kept a back up of everything on an external drive so i can afford mistakes.
couple of quick questions:
Is there anything i can do prior to using mbp to improve accuracy?
when ive looked up and scannded all the unclustered files, what happens to them? Do they stay in the source folder or are they all sent to the rh pane unde a particular heading and if so, does that get saved as a folder?
Hmm, not really, I guess you could move things into folders and update album tags, but that’s all stuff you can do in Picard anyway.
Let’s run through the basics quick, I think that’ll help. You might be aware of some of this already but I’ll just cover it anyway.
Left panel:
The left hand panel is your local files, which haven’t been matched to anything in the database yet. Clustering them groups them into albums based on the album tag or folders they’re in. You can drag and drop files in and out of clusters manually if needed. If using lookup get your clusters right first (can help to work with less albums at once).
Right panel:
The right panel is data/releases/tracks in the MusicBrainz database. The aim is to get all of your files on the left matched to the correct releases + files in the database (see methods below). You can drag and drop your files from left to right and vice versa, and drag and drop them between songs and albums on the right. Nothing will change in your files unless you hit ‘save’. You can right click a release on the right to view and select other ‘versions’ of the release.
Populating the right panel with releases:
When you select a cluster or a file in the left hand panel and click ‘lookup’ it will search MusicBrainz for a matching album based on the existing tags (e.g. artist, album, song titles). It will do its very best to keep clusters together, as it assumes a cluster should be a single album, and will try find that album in MB.
When you select a cluster or a file in the left hand panel and click ‘scan’ it will search AcoustID for the files audio fingerprint. It ignores all existing tags. (I’m not sure if it ignores clusters as well?) It is great if you have crappy existing tags or no tags/a mystery file. It will throw up a lot of compilations etc, as you can imagine, since it’s ignoring the album tag and songs can be on heaps of albums.
A third way of populating the right hand panel is clicking ‘lookup in browser’ in Picard. Now when you browse the MusicBrainz website little green ‘tagger’ icons will appear on release pages. Clicking these will load the release into Picard, on the right hand side of course. You canl then drag your cluster or files from the left panel onto the loaded release on the right.
Let me just see if i understand correctly.
You put all your music files in the left hand pane.
The aim is to sort them all into the album they come from (lets call it the original album).
Some will be identical tracks coming from different albums and where MBP puts those tracks in different clusters/albums you can put them all in the one original album manually by dragging and dropping a file in the left hand pane into another cluster/album folder also in the left hand pane.
Alternatively, you can tell MBP that a track which is, for instance, in a compilation album is actually from its original album by manually dragging the file from the left hand pane onto the original album and the specific track in the right hand pane.
I’m going to pause there to check ive understood so far…
You can, but do you really want tracks tagged to a different album than the one they came from? The answer may be yes for you, but generally speaking Picard is best at dealing with full albums.
e.g. Picard functions best if you match/tag all of a compilations tracks as that compilation. And assumes if you’re wanting to tag the original album you’ll have or get the original album and then tag the whole thing.
But yes overall you are correct, you can drag and drop all over the place until you’re happy!
Thanks - I’ll have a think. I have alot of compilation albums of individual groups, e.g. Uriah heep plus a lot of single uriah heep singles from other albums, but not the whole of the album they come from. They’ll all end up in different album folders but under that artist which would be okay. But what about tracks that come from other artists Now Thats What I call Music and so on. Will they still appear under the artist or will the artist be ‘NTWICM’ etc? For instance, would a cliff richard track on a now album appear in the cliff richard folder or the NTWICM folder?
Or is the idea to maintain the integrity of albums and their tracks as much as possible and let your music sofware, MWP or whatever, to take care of categorizing them?
Those albums will go into a ‘Various Artists’ folder
Depends on the person, but that’s how I approach it!
All the tags given by Picard give you quite a bit of control in a good player (I use foobar2000). For instance, I wouldn’t manually move all an artist’s songs into one folder, because I can just sort or search the library by the artist tag anyway.
I don’t keep loose tracks from albums, but maybe you could do something like - save and move all the full albums (‘gold’ albums in Picard). Then go into settings and change the script for where to move the album. Change ‘%album%’ or whatever to ‘Misc’ or something. Now save what’s left, and they will move into a ‘misc’ folder in the artist folder. Just an example of what you could do!
You have a backup so you can play around which is excellent.
There’s two main tags for “artist” one is Album Artist, and the other is Artist. In the case of a various artist compilation, the Album Artist will be “Various Artists” and the Artist tag will be the actual artist. A common way to organize is to have the Album Artist as a higher level folder name and the Artist as a part of the filename. So you might have:
…\Various Artists\Now That’s What I Call Music\03-Cliff Richard-Devil Woman.mp3
