MediaMonkey forum

I have an idea that I'll admit most people won't use, but it could be very useful for a lot of MM users and MP3 collectors. Now to describe the problem that this feature solves. I have many different versions of the same song. Of course, MM doesn't know that these songs are alternate versions and treats them as different songs. This means that some songs have a higher chance of appearing in shuffles than others due to sheer amount of versions. My request is to implement some way (maybe a tag of some kind) to group together multiple files that are different versions of the same song and let them share the same shuffle chance.

For example, let's say you have these 6 songs in your library. In MediaMonkey currently, random shuffle will give each song a 16.66% of playing.

16.66% — Big TV
16.66% — Big TV [Demo]
16.66% — Big TV [Live at Hammersmith Apollo]
16.66% — Fire and Wings
16.66% — Holy Ghost
16.66% — Take It Out on Me

But with the ability to group song's shuffle chance together, you could make the shuffle probability look like this.

*11.11% — Big TV
*11.11% — Big TV [Demo]
*11.11% — Big TV [Live at Hammersmith Apollo]
33.33% — Fire and Wings
33.33% — Holy Ghost
33.33% — Take It Out on Me

I think it would give users more ability to enjoy and clean up their database a bit more so that alternate versions don't completely crowd the database, but still appear now and then.

Not a bad idea (even when the sample calculation goes beyond 100 % probability ) but maybe difficult to implement.
I am trying to work around this issue by using auto-playlists. Every song that is a version of the original (= my favourite) gets tagged "doublet" in the user defined tags. Now I have auto-playlists that exclude all songs with "doublet" and others that don't.
Not exactly what you want to achieve but maybe a step in this direction.