These days, with 24bit recording, there’s just no need to compress whatever it is you’re recording. Back in the olden days, with tape, there was a need. You needed to compress the signal to help fit your audio in that dynamic range of tape between total tape hiss and overload (you could also ride the level manually if you weren’t lazy, but most of us were).
We now have over 125dB of dynamic range available to use in digital recording. As long as you set your levels not to clip at the loudest moment of recording, you are probably going to be all set.
Now, I did say there’s no need. But I’m sure there is plenty of “wants” to record with compression, say for a special effect, or if you know you wanna squash the daylights out of your drum room mics, or if the vocalist knows how to sing to a compressor, or if you happen to have an old Fairchild sitting there, by gosh, go for it.
BUT, I really don’t recommend it. The reason being is that YOU CAN’T UNDO IT.
For example, you just tracked an acoustic guitar through a nice 1176 compressor. The take is great, the player is really happy with it and you close up shop for the night. The next day you come back for a listen, and you notice with your fresh ears that to your horror, every time the guitarist plays a thumpy low dropped D with his thumb, the compressor pumps and it brings down the sound of the ringing harmonics of the two high strings. And the guitarist doesn’t like it. Guess what? You gotta either live with it or record it again.
So what should we do if we really like our 1176 on our acoustic guitar?
The solution is to track it without the compressor. Once you have the take, then simply patch out of the DAW, into the 1176 and re-track the guitar thru the 1176 listening carefully to make sure that pumping isn’t there this time around. It’s digital, not analog so we’re not really losing fidelity here (some will argue this, but to my ears, it’s fine).
I do this with everything these days. In fact, I rarely EQ anything to “tape” either. Everything goes into the DAW flat, with no compression. Just a microphone and preamp, straight into the computer. When it comes time for mixing, I can make a better decision on how much or little I want to EQ or compress something.
This is also hugely helpful if the record you are recording is going to be mixed by someone else. No sense in painting them into a corner with what you thought was a great compressor and compressor setting, when they may have a much better idea of what needs to happen (and they also may have better gear).
I have mixed many things over the past few years that have had so much compression on them, it’s impossible to really change anything. It particularly makes me crazy on vocals. An over-compressed vocal gets very “Ssss-y” and doesn’t take well to either EQs or de-essers.
If someone really wants to track with compression, I will happily put a plug in on the track to approximate what it might sound like in the mix. Sometimes this helps, especially with singers.
And finally, this always seems to get swept under the carpet: dynamics are actually really great. We could all try harder leaving them in as much as possible.