|
Wow, this thread got meaty.
Look, vocals are not "the most compressed" element in a mix.
Pop, in general, does use a fair bit of compression on vocals, but I'd be hard pushed to name one record where the vocals are THE MOST compressed.
Rock can vary in terms of compression on vocals, as singers often have serious crescendos, but in those cases it's more limiting, and the ompresison will range from light to heavy, but again, I don't think I can name one record where the vocal is the most compressed (unless you're talking a specific effect like distorted and over compressed vocal sound for a certain aesthetic).
The whole thing about desk bus compression has gone way too far and IMO, blown out of proportion. for every famous record that is saturated I can name one that has little to no compression (Thriller Album has virtually no compression at all, not even on the drums).
That late 70's rock sound was the result o7 desk compression, but it was certainly not the norm - it was a fad led by a technological breakthrough, that everyone jumped on at the time, like the wall of sound technique or vox amps on the beatles sound etc.
In terms of tracking, yes, you some compression but how much depends on two things:
1, if it's a particular aesthetic you're going for, which is a matter of taste (and not pertinent to this discussion)
2, Technique of the singer.
With No.2, engineer often find themselves dialling for more compression when they've got someone with bad mic technique or low vocal control. Rappers nearly always need a fair amount becuase they just shout at a mic and you can't dial in for dynamics when their forte is 20-30db louder than their spoken parts....so bear in mind this is technically a band aid to fix bad technique.
For most of the projects I've been involved in, tracking compression for vox is minimal and subtle and mix comression is the same, if any at all. In fairness, it's mainly film work but we do commercials as well but again, we prefer to go easy on compression, and leave that the post processors to squash the shit out of it if need be.
Ideally, I would work so that I have both compressed and uncompressed signals going to tape. Reason being, is that sometimes you need a little "performance" compression (like adding reverb to a vox for the singer to hear) and just enough so that if they spike you're rounding it down, but other than that, you want to keep it as real as possible.
IMO, if you're having to add more than that, then it's either to fix a problem with the singer or to create a specific effect, but certainly, neither of these should be the defacto way of working or the "norm".
Last edited by DJ RANN on Nov-12-2011 at 20:40
|
