I'm wondering how many people set the gain of each track before beginning the mixing process so that the instruments are all around the same loudness (lets say 0DB for the sake of this discussion)?
 
I saw it in a tutorial and it seems like a great idea.
 
Thoughts? Best practices?
 
Thanks!

Conversely,  I'd call it a really bad idea to not check the raw track levels are reasonable by adjusting the gains before you start mixing...

For me, gain adjustments occur at every stage of mixing from beginning to end.
 
At the start, it's a way to initially get all your tracks in the same ballpark. At the end, it's a way to make minor volume tweaks to tracks and synths that have been frozen or that have extensive volume automation on them, or to adjust reverb and distortion bus levels. In between, it's used to level vocal, bass and drum tracks.

I do gain staging right from the input volumes on my audio interface, and using the master volumes on individual presets for soft-synths, so that I don't need input trim adjustments or initial fader adjustments to have all my tracks coming in initially at around -15 dBs, with Sonar faders at unity.
 
That puts me in good shape from the beginning, and I always end up in good shape when it's time to mix.
 
Others do things differently, but for me things work out OK.
 
Bob Bone
 

only do that in case I can't move fader anymore, which is not often.
then I use input gain, clip gain, output gain or volume
I don't think you HAVE TO set gain input to get a good mix, it depends if it helps in your workroutine.
Suggest to try yourself to see if it makes things easier/better..
 

I just adjust for the green glow(2i2) with NO orange on the hardest attack(guitar,bass, and vocal). It seems to be working fine. It seems to fall in the -12 to-16 ballpark on avg.

Excellent answers! Thanks everyone!

remember the basics.
 
even just raising a channel eq a bit, raises gain.
 
so as bit said, 
it's a constant correction.
 
fx's drives gain....
panning can drive gain, depending on your chosen 'pan laws'.......
 
ultimately, i think the best case scenario,
is where you've recorded your individual tracks with enough headroom,
that if you zero out all your individual tracks faders,
and your master and sub busses are at zero (unity),
then you mix is almost finished!
 
 
 

I keep getting confused on the back and forth between the terms unity and zero....
 
Question - When you say zero out the faders, my mind first goes to bring them all the way down. But later there is a reference to zero(unity). If that is the case, what is the term for physically dragging them all the way down?
 
Tx

I make sure when I start mixing that I get the peak level to sit between -12 dB and -18 dB. I like to treat -12 dB in digital domain kind of like 0 VU in the analog realm. I've noticed that some plug-ins can react negatively if they're dealing with too hot of a signal. I also try to maintain that level in between plug-ins as well. A lot of it has to do with feeding the plug-ins a healthy signal. This really helps the processing, the headroom, the mix in many ways. The faders at the end of the channel do end up moving a bit but I feel as though there is more control/intention with how they end up moving.
 
I totally agree with what batsbrew says about the best case scenario. It does set you up for an ideal mix situation.