Automation vs. separated channel mixing?

Question

I have a mixing question I've been dying to ask.
When it comes to mixing, a friend of mine swears by automation.
His technique is to put all lead vocals (or any similar sort, so he would put verse 1 & 2 and bridge lead on one channel) on one single channel and automate the level of that channel up and down according to the part of the song and obviously the volume he wants from the vocal part. He has left me with the impression that "That's how everyone does it".
I on the other hand have never bothered with automation when I mix a song (speaking primarily of vocal channels) unless I really have to. I simply just use more separate channels even if it is the lead vocal. So for example if it's verse 1 there are three parts of the verse that I want in different volumes, I just put all three on different channels in logic and copy the COMPRESSION, EQ setting and bus sends on each. So in the end I usually end up with between 5 or 20 audio channels just lead. I can then also obviously set EQ and COMP and whatever else I wanna do on each channel individually as the need arrises.
This to me is way faster because while mixing the entire song I can quickly get to the channel that needs adjustment and change the volume according to what I hear.
Mixing with Automation in the described scenario by my friend is just clumsy to me and harder.
Please let me in on your preferred way of mixing.... with automation and similar to the way described that my friend does it or do most of you also separate audio parts in to various channels as the needs arrises?
Of course I am always willing to learn.
Thanks.
Message was edited by: Chance Harper

Answer 1

I mostly use similar approaches to the very well described methods in Bill's response - except that first, I spend a lot of time (where necessary) to tweak levels within the vocal tracks in the sample editor. In there, I'll first remove pops, thumps, noises etc. I'll then change gain on individual phrases and sometimes even words or parts of words. I don't use compressors to replace riding a fader or automation.
Then, once my tracks are cleaned up, I'll often use two compressors - one dealing with undesirable remaining peaks, and one dealing with RMS levels, but mostly for the sound it gives me because of the choice of compressor, not gain control. All of this apples to other instruments too - my goal is always to get the raw tracks as close to 'right' as possible before applying anything - which goes back to the recording stage, obviously.
As I mix OTB often, automation I use is mostly to push-pull in fairly coarse increments. Using the same 'parallel' compression routing, but not actually using compression, I might apply affects to another bus, so I'm generally mixing the affected vocal track with the untreated vocal track, and sometimes swapping them for more extreme effects. As I'm often OTB at this stage I don't have a lot of options as I use the creaky old (now obsolete) Uptown automation system - or none at all on certain tracks. Somehow there's a magic that happens when there are human hands on faders during a mix, and the mix becomes another performance.
Of course, this applies mostly to the music type with which I work - roughly 80% audio, 20% virtual instruments, so everyone's techniques will be different.