Tonight I spent some time levelling some pre-recorded stems against pink noise. I normalised their peak value to 100% and then moved the faders until I could barely hear them or at all. Then I thought fine, now I have all the drums, synths and vocal samples levelled - how do I maintain their volume difference level and have them all move in unison. 
 
So after a bit of research I found grouping was the answer in the console (or sequencer tracks). I created 4 groups of sequencer tracks. Now this is probably the first time I tried using Sonar like a proper mixer, however I need to know one thing. Do the tracks still move when automation is applied or is some other further processed required to make this happen. This important as I need to maintain the balance between each track during the mix process. 
 
I will probably use a baked stem bounce for each track group at a further mix stage. I suppose I am wondering if the behaviour of Sonar is like an automated fader approach rather than individually tweaking each part like I have in the past. Or maybe I just hit on a new approach to the mixing in Sonar.
 
The differences between each track are subtle like -2.6dB and what have you, although this should help me get more accurate and balanced results. Spending more time creating rather than fixing my mix.  It's more enjoyable when you know things are just going to work. 
 
Thanks for any tips and help. I realise this is not a straightforward process and just for your information. Very little effects will be used other than delay or some spatial effects. The drum samples are mono mainly, the symth pads are stereo and all the vocals and part effects are mono. 
 
Again thank you in advance.
 

It seems so, awesome this will save me a lot of tweaking later on. This is probably obvious to some of you however I haven't really being doing much music stuff. I let my license expire and had to resubscribe. 
 
Long story I'm learning how to use this program properly now.
 
 

look up Fletcher–Munson curves

I'm not sure that helps me. The fundamentals of frequencies and perceived listening levels are not really answering my question. I sort of have a grasp on that although you can't know everything.

I answered my on question though. However I was looking for more other tips to the mixer automation rather than add more science.

Thanks though batsbrew.

Sonar allows you to group tracks as you stated:
https://www.cakewalk.com/Documentation?product=SONAR%20X3&language=3&help=Mixing.41.html
 
 

i guess what i was trying to say,
is that normalizing by level,
is not going to get you to a 'balanced' mix.
even with pre-mixed 'stems'.
 
gain staging, and mixing 'in combination' with arrangements and frequency ranges,
that kind of thing..
is where the fletcher munson thing comes into play, 
and suddenly 'normalized' levels don't really matter any more.
 
 

Ok....I'm with you. Have an instrument graph although my current thing is mostly electronic so paying attention to FFT frequency analysis. Loudness meter and that on my master bus at the moment..

Out of curiosity, is there a reason that busses will not work in your situation?

I would be hesitant with your approach, since fluctuation of how tracks sit against each other (even on the same buss) is an important dynamic to consider.

It's an exercise for me. I will be bussing them however I want to be able to balance each mix group by ear and not always rely on compression. Also gain or cut in volume should move up and down at the same time.

Individual fades can be done on clips.
 
Normally I just plonk a compressor/eq/de-esser or whatever on a track. Group buss it however I spend some time altering some stuff later on. So this might be more thorough than doing post-production however will save me time later on.