Wow, I'm not sure I follow exactly what you are doing. Different parts, recorded in different places at different levels, in different sounding rooms. And now trying to blend them together into one piece of music. That's a lofty goal. i agree with taking the time to volume automate. i wouldn't use any normalization or compression to deal with volume issues. Stick with the volume automation. I would only use compression on a mix buss to blend any transients. as mentioned earlier in this thread. This however is a great post as to the importance of pre planning prior to recording. This is one of those things you do once so you know never to do it again. again, if im understanding everything correctly. Good luck and please post it. would love to hear.
Yes, It's my biggest "engineering" task yet. I'll give you a quick example of what I'm up against.
4 vocal parts, lets call them Low, medium Low, Medium high and high, with 1 panning left, 2 centre and 1 Panning right.
Then, in some cases the parts will be center and not panned.
In some cases there will be an extra (5th) part.
Then, some of the multipart harmony's need to overlap another section of multipart harmonys, or be panned, or have the level played with.
.. and of course there is a main vocal, that during sections will make up one of the harmonys.
So to try to organise this into your standard 1 main Vox track and 4 harmony tracks just won't work, yeah? There's probably a few ways to divide this up in your DAW and manage.
Now, because I'm not a singer, the very very high parts are precarious (to say the least). When I nail one good take that's in time and in tune, it STAYS because statistically it's the best that's gonna come my my throat. I then have to deal with whatever this turns out to be. It may be soft or varying in volume; far from say the mid range parts where I can belt it out. Same problem for the low notes.
So now (or very soon) I'll have to try to manage all the above into some sort of system so I can get to the end, ... you know and win the race and collect all the gold.
you know, the easiest thing to do, is to cut to the chase
do vocal rides, and if you have the ability to do automation (what modern DAW doesn't do volume automation) simply mix it.
An update on the big project.
i have now (finally) finished all vocal tracks. There are ... a lot. I'm having my first ever trouble with channel level - I'll explain.
With so many vocal tracks doing different things , I thought I'd double up on the uses yeah?
So I have one group of 5 part voices, double tracked - so 10 tracks. And these are sent to a single group track so eq/compression etc can be managed. (Sounding good btw).
But for another section of the song, I have 2 parts. So I thinks, why not just put these two parts in two of the 10 tracks (mentioned above)? Because (duh) the level of 10 voices singing is way more than the level of two voices singing. All of the group input/output levels set for your eq and compressor work for one section pof the song, but not another.
So I either go and reduce all of the 10-track levels to make it equal to the 2-track level, or vice verca. Or I just leave the 2 tracks where they are.
Again, resources are plenty so I can't see (at this stage anyway) why I shouln't just create extra tracks/groups and use them as necessary.
This is just one example of what I'm currenly sifting through and already I feel way out of control. Still, what's the worst that can happen - if it all get's too hard, I'll go back to a previous version and try again.