Great find. I was wondering to myself if it should* be done for pre-mastered tracks. Was often hesitant not to compress the vocals too much but the instrumental would always seem to fight the vocals down.
What do you think about sidechaining the vox to the beat? I often get beats that are "mastered", usually squashed to the max (about 90% of the time lol) and I've found that it helps to get a little more dynamics and helps the vox stand out a little more. I'm talkin very light, so the when the vox peak they will bring the beat down justttttt a little bit.
I do a lot of these. I’ve gotten a reputation as the person to go to when you are stuck with a 2track beat (beatmaker lost the files, computer crash, etc…. sh!t happens).
I use a transient designer type plugin to undo (within reason) the limiting and 2buss compression on the beat. That with some very careful m/s EQ before and some shaping m/s EQ after to get the right overall balance on the beat can generally get me where I need to go (within reason, a bad mix on the beat is still a bad mix – you can shine a terd, but not polish it). I’ll sidechain the “all vox” buss from the beat. As someone mentioned earlier, it’s VERY light (couple dB max). It mimics what the 2buss compression would do to the vocal. Gets the vocals to sit nicely and sound like they aren’t dropped on a compressed 2track beat. Usually I can get things working enough I can even go and add just a little 2buss compression of my own on the final mix.
The hard part is really the transients and the EQ on the beat, and then getting just the right touch of sidechain compression. Once you get that right you can pretty much completely ignore the fact you are mixing to a 2track compressed and limited beat.
IMO it’s very very difficult to mix vocals to a compressed and limited 2track beat without sidechaining. Doing so requires making other comprimises and usually you can feel that something isn’t right. But with sidechaining you can COMPLETELY fool people.
I’ve got one client working on an interesting project where he’s dropping his own verse into commercial records and you honestly can’t tell on any of the songs that he wasn’t right there in the studio with the other artist(s).
What I'm gathering from reading the original blog post and the comments is that transient processors are used to accent beat hits. Am I correct? I'll still search the forum because I know they have the thread, but I want to hear you guys' thoughts on it from our conversation here is all.