Heya. I just saw your pretracking thread, may as well toss in my $0.02 on this one too.
I have a 3 piece rock band so it's simple really. I don't edit I just start over if I screw something up, but our stuff isn't very hard to play (vocals on the other hand...). When I do vocals and am having problems I go 1 song segment at a time, like do a whole verse, stop, gather myself, do a whole chorus, stop, gather myself. I can't deal with doing every line separately, it takes forever, bores me to death , and sounds wierd.
Making a mix of a 3piece band for me is easy enough if the work is done at the tracking stage. I'm sure it would get more complicated with more stuff in there, but I just have the bare bones rock trio. I have 2 tracks of drums (just a stereo of the whole kit, so my mics are already EQd in the submix) 2 tracks of bassline (DI and amp mic) blended, and a stereo mix of guitar (one through an amp sim and one amp mic - a new trick I've been messing around with, blending more than one guitar sound just to experiment making it fuller toned.)
I start the sliders in the middle. listen through all of it. drop out EQs if a certain tone is too much. Keep vocals blended, because they naturally stand out from music because of the sound of words, they don't have to be at a high volume to cut through a mix, same with guitar "leads", they stand out without being turned up because they are naturally noticable.
I like to have my bass part big, but I EQ it all so it is tight sounding big by getting rid of quite a bit of mid and some low, instead of too boomy wompy sounding whatever you call it. The guitar pretty much just gets left alone, the tone is just how the amp/sim is set and it gets tracked that way, as do the drums because I EQ them as I track since its a submix not an individual drum.
I drop the low EQ out of my vocals so they can be turned up a bit without getting boomy too, but I try to do as little as possible, and nothing extreme.
EQ should not be obvious, it just gets rid of stuff that muddies the mix, or cuts through harshly, rather than adding stuff. All the tracks get sent to a tiny bit of soft toned stereo reverb at the end because it helps smooth it all together, then I compress it, not much, just enough to make sure it doesn't spike digital distortion and all that. I guess that's "mastering" but to me its a start-to-finish process, not 3 separate concepts. This is a process for a very small band. If I had to use a second guitarist I would think that pan separating them a bit left and right would be a good place to start so they are two distinct parts.
This is of course just my amateur newbie take on it, but I'm curious what you do as other trail-and-error techniques may occur to you too.
