So if you are turning the gain knob (up or down) are you adding or multiplying?  And apologies to Jeff if he explained this in one of his very detailed posts that I skimmed.

multiplying. Input with a gain of 2 = input x2.

batsbrew
i think ultimately, the IDEAL situation is that you get your gain staging right from the beginning..
by tracking conservatively...
and then, when you take all the faders across the board, up to ZERO,
you should be peaking exactly where you want to, on the master bus.

There are different approaches to getting a final mix that is right and the complete mix is also hitting the ref level of your choice, peak or rms. For me it's all about rms.
 
Both balancing methods will get you there and they will sound very similar.
 
Having all your faders at the same level and at unity and creating the right mix and signal level requires the rms levels on the tracks (or somewhere else)  to still vary in order to get that mix correct.  How and what do you adjust in order to balance groups of instruments which is essentially what mixing is all about?  A bunch of faders at all unity won't give you much visual feedback as to what the mix is like.
 
I prefer to keep all the rms levels on the tracks the same and then build the mix using the faders.  Some will be at or around unity but many others will be in various positions in order to create the balance.  Some very low.  The faders in this scenario will give you some idea now what things are loud and what elements are quieter in the mix.  This is more how it used be with analog multitrack tape.  The tape tracks were all at the same level (rms) and the mix was built on the console.  With practice you can get a great mix where everything is balanced and the whole mix is sitting at your chosen (rms) ref level at the same time.  And the VU's are moving just the right way too.
 
Tracking your tracks say down at -20 dB FS (rms) is very conservative.  Nice level with 70 dB or more of dynamic range below that and 20 dB of headroom above that ref level.

I THINK it goes without saying,
that when i say IDEAL, that would mean you were paying attention to things like RMS while actually tracking,
and knowing in advance, how you want to mix it.
 
that said, my faders end up all over the place!!
LOL
 
because i almost always end up carving some things up eq wise,
and using raw tracks here, completely limited tracks there,
so the faders go where they have to go.
 
point is, if you are tracking so aggressively that all your faders have to come down to about -14 just to hit -6 peaks on the master,
you're too hot.
 

I use Hornet VU Meter to normalize the RMS values of all tracks. I put an instance on all tracks and set to -18 and enable Auto Gain. Then, play the song. Each VU instance will set it's Auto Gain to the number of gain adjustment each track needs to RMS at -18. I don't like to keep them on the track, so then I adjust the Gain knob of each track in Sonar to match the value recommended by Hornet, and remove the Hornet plugin from all tracks.
 
to me, this gives a very good starting point for the mix and requires little volume adjustments with the faders, rather than having a project with faders all over the place.
 
After that I use the Waves Dorrough meters 'cause the be awzome.

Hi, I have a tendency to record a bit hot when using my tube preamp, to get the grit I want. I then use the gain (knob) to adjust to desired level when mixing (ProChannel overload etc). But that is just me.
All the best.

What is right about Bat's post I think though is that if all your tracks are at the right rms ref level and you push up each one, one by one up to unity (eg solo) what you should see is 0dB VU on the masterbuss. That will tell you how accurate your track levels are.
 
Zargg71 I would imagine any grit you are hearing is coming from your tube preamp and that is the only place where the grit is created.  That is a technique left over from a previous time when we did drive mixers and multitrack tape machines harder in order to add some crunch.  Now though the grit is not related to your digital recording level.
 
So your final output from your preamp can be recorded at any digital ref level eg -20 dB FS or -14 dB FS or even -8 dB FS. It should sound exactly the same.  The only difference here is how loud those tracks will be.  So why not record any preamps at your chosen rms ref level instead.  Then you will have your tracks pretty well ready to go and they will be at the right level before you mix.  No changes to that track level will be required before you proceed any further.
 
 

Jeff Evans
 
Zargg71 I would imagine any grit you are hearing is coming from your tube preamp and that is the only place where the grit is created.  That is a technique left over from a previous time when we did drive mixers and multitrack tape machines harder in order to add some crunch.  Now though the grit is not related to your digital recording level.
 
So your final output from your preamp can be recorded at any digital ref level eg -20 dB FS or -14 dB FS or even -8 dB FS. It should sound exactly the same.  The only difference here is how loud those tracks will be.  So why not record any preamps at your chosen rms ref level instead.  Then you will have your tracks pretty well ready to go and they will be at the right level before you mix.  No changes to that track level will be required before you proceed any further.
 
 

This was a reference to when I use my tube pre. I use both the tube pre and tube out (LA610) to get where I like it (if I need a bit of harmonic distortion, and I often do), hence the hot recordings On my other pre, I record approx -12dB.
All the best.

recordings typically are not 'hot'...
 
they are either clean, or distorted.
 
tracking, is not the place to drive volume.
that'd be for the mastering phase.