Do acoustic instruments need compression?
Yesterday's Record Producer Daily on mastering brought up a couple of interesting comments.
One from Remy Ann David remarked, "I've been told numerous times that my mixes already sound mastered".
This is of course exactly how things should be. When the mix leaves the studio it should already be the very best it can be. If there is the merest inkling in the engineer's mind that mastering will be able to improve it, then the mix simply isn't finished.
Another from Alcohol of Massachusetts commented, in relation to recordings of acoustic isntruments, that, "To make the recording more suitable to listening, some compression is necessary to tame the dynamic range."
Yes, I agree with this. If you record a group of acoustic instruments with just one stereo pair of mics, then the dynamic range that is captured will be uncomfortably wide for domestic listening. And iPod/in-car listening too for that matter.
Without disagreeing with this point in the slightest, it does raise for me a couple of issues that I would like to talk about.
The first is what happens when you add more mics on top of the simple stereo pair.
As you know, if a group of acoustic instruments or voices sounds good in the natural acoustic of a room, then it is possible to make an excellent recording using just two mics in the coincident crossed pair, near-coincident crossed pair, ORTF or spaced omni configurations, according to your preference.
All it takes is to find the right position for the mics. ("All it takes"! - it can take ages of trial, error and adjustment to get this right.)
But there are reasons why you might want to add more mics, even in a good acoustic.
One is if you are in a hurry.
Is there such a thing as a hurried recording engineer?
Well there shouldn't be. Enough time should be booked to do the job properly. But in broadcasting there often isn't so much time. And in television recording there are the requirements of the cameras to consider too.
To get a good balance more quickly, you can set the stereo pair closer than you normally would. Then add two spaced ambience mics towards the rear of the auditorium.
The outputs of the pairs of mics can be balanced on the faders to give the optimum blend of direct sound and revernberation.
Many would say that this isn't the best way to work, and indeed it can result in the reveration not sounding entirely 'connected' with the instruments.
However, I for one rather like it and I would sometimes choose it for preference over the simple stereo pair.
Another reason for adding more mics is if the ensemble is large. What can happen here is that the instruments that are remote from the stereo pair sound dull and distant in comparison with those at the front.
One solution that is often employed is to add more mics for the rear instruments. You can go further than this and mic every section of instruments individually, plus the main stereo pair.
This will solve the perspective problem, at the expense of possibly a more confused sound.
But something else happens too...
The dynamic range is reduced.
I remember when I first became aware of this when it was demonstrated to me some time ago by Mike Beville of Audio & Design.
In my own recordings since then I found that it was most definitely true. And the more mics you add, the more the dynamic range is reduced.
We are not talking about major differences here - around 6 dB at the most I would say. Still, this is enough to make an audible difference and make a recording more listenable at home.
There is another point I would like to make...
I believe with a passion that compression is not the best way of controlling dynamic range.
This may seem odd since that is the whole reasoning behind compressors in the first place.
The fact though is that compressors were invented for the broadcast industry. And as I have said already, the broadcast industry works at a fast pace - they need compressors. But recording engineers outside of broadcast should look at another option first.
And that option is to control the dynamic range manually. It is as simple as this - raise the level when it gets too quiet. Lower it when the level comes back up again.
This used to be done completely manually on the faders, but now it can be done more conveniently with fader automation.
When a musically-aware engineer adjusts the dynamic range of a recording manually, the result is far better and more natural than can be achieved with a compressor.
Compressors do of course have their place. Reducing dynamic range on a very short-term, tens of milliseconds, basis is one. This is useful for enhancing a popular music vocal.
Going back to the original comment. Yes I do agree that an acoustic recording will need some reduction in its dynamic range. The engineer should choose whichever method gives the most natural sound.