View Single Post
Old 13th July 2017
Originally Posted by mattiasnyc View Post
If Bob used the term the way it was otherwise used then stems are a subset of the audio that constitutes the final mix. It's important to make that distinction. If I'm not mistaken they originate in post production which is different, and any set of stems should recreate the final mix. This means that if you send your individual drum tracks to a bus, do processing on it, then after that point it can't change. Some people might render that "stem", or consider it a "stem", but still do things to it afterwards, even just change the level, and at that point it's no longer a stem unless you re-render it after the level change.

Grouped/bussed tracks that aren't as "set" in the sense that they recreate the mix as is would be called "splits" instead of "stems". So when I receive music from a composer for a TV show or film then from his perspective he might call them "stems" because when I add those together without changing level or anything they'll recreate his music mix, but to me they're "splits" because my final mix is the actual TV show or film. My stems would be dialog, music and effects (for example).

So I've disliked the usage of the term "stems" for a long time, especially in the music field, because people don't seem to know what it means.
Katz's Mastering Audio defines stems as, "...a special kind of submix. For example, there could be a lead vocal stem and an instrumental stem, which when summed equal the full mix."

I think this has been the accepted definition in music recording/mixing/mastering for awhile. Your definition for stems having addition content to the composer's stems doesn't seem to be a conflict to me, you've created a new mix, so your stems are your stems. Maybe differentiating "music stems" vs "audio for video stems" or something. Either way it refers to submixes that when summed equal the mix, just at different stages.