Too much compression? Advice requested.

[slopeshoulder]

Hi:
I'd like to get advice on where to draw the line with compression. It seems like I could put comps on a drum kit pieces, overheads, room, pzm, and group, then on bass, vocals, background vocals and guitars and keys, then on groups. Then maybe on the outputs, then at mastering. Last time I had a record mastered (mixed by a pro BTW, not by me), the mastering engineer added a lot of expansion and also told me that 90% of the compression he was doing came from passing the mix through tape. Hmmm...

Doesn't all the possible compression (tracks, groups, 2 bus, mastering) sound like too much? Why not add a little to the kik and snare, maybe a little room, and then group OR overheads? Vocal and solo? Then just let it be if it sounds good? If the performace dynamics of the parts are just right, won't compression change those just right dynamics? So: fix, effect, and glue, but no more, right? (I don't record pumping hip hop or dance. I don't listen to radio)

Extra babble:
I'll be doing my own mixing of a lot of material soon (4 albums tracked, waiting for mix). The good news: we can do this. The bad news: we don't know how to do this.
My answer:
- listen like crazy and study like hell (I have and continue to)
- use reference CD's
- use plugs judiciously
- use plugs to fix, for effect (just passin' through), nd for glue, but not just because I can
- stay the hell away from the main outs (unless for color and low settings), leave that to mastering (elsewhere)
- get and use plugs in a way that reflects what i want to hear and what people I respect seem to like; set up a basic environment that suits the project (fopr example, maybe Chandler TG on drum room, 1176 on snare, LA-3A on overheads and kik, plate on snare and vocal, Neve comp on drum group, API EQon drums, API or Neve EQ on electrics, or Oxford and Sonalkisis and Aireq where they "should be", Pultec baby Pultec!, etc)
- realize that life is short, the time is now, perfectionsism is folly, and the mastering engineer can help.
- mix for people, not for gearslutz (sorry)
- wait until I get answers to these questions:
1. if the musicians play with dynamic and feel and artistry (they're me and I do), then I shouldn't have to be riding faders a lot, right?
2. if I do ride faders, I can compress less, right?
3. if the recorded tones are excellent (they are: great gear, good ears, pick tones and play to fit what's already recorded), I should only have to EQ for the occasional correction, to make things fit in the mix (usually subtractive, and some sidechain between kik and bass), or for effect, right?

This is only a set of rules of thumb. I don't expect a personal mixing tutorial here. But do I seem to be in a good place as a place to start?

Input and advice MOST welcome. Thanks. All he best.

[james.ca]

Hello
Try it both ways and see what results you get.
Half the battle with compings how much and not to hear it being used .
If you have the band to work with try a bunch of things
You will come up with your own methods that's what a good producer does anyways .
All you want is to have some thing the mastering engineer can work with the better you give the better his master will be !
It's not about right wrong it about sound quiality . How you get there doesn't matter as long as you get there .
I'm back inrecording school a swell tryng to fill out bunch of voids as well never to late to learn
James