Stems - a conversation
Where I work we have a handful of audio engineers that each have a different idea about what stems should be, how they should be used, and how to create them.
It is my opinion that the stems should equal the mix. In order her words, if I bring the stems into a session and play them all together it should sound exactly like the full mix, and measure exactly the same on the loudness meter. The way I am doing this, prevents me from doing any master bus processing. I limit each of the sub mixes for each stem, then combine them to a record track which I then measure with the loudness meter. This also allows me to record the mix and all the stems at the same time.
Other engineers prefer to mute everything, but one stem and bounce out the mix several times. this allows them to do master processing, but the downfall is that the stems do not equal the mix.
Another engineer will bypass all of the mastering, and just bounce out each of the stems from the sub mix only, assuming that it will come back later to be mixed and mastered if something changes.
So my question to you all is: what is the industry standard? How do you all handle stems? Do you assume any changes will come back to be remixed? What are your thoughts?