I wrote a long entry, Internet Explorer froze up, and I lost it.

Have I mentioned lately that Windows is a piece of crap?

Anyhow - back to our regularly scheduled post.

CHOIR, PART 1

Last Friday, Pat (aka Tony Alva) recorded the Taize Choir at a Catholic High School in Georgia as part of the ongoing partnership between Smoke and Mirrors and Pat's Grey Cat Sound. Originally, Ted (aka Jackson) was scheduled to make the trip down to assist, but at the last minute he got an emergency call from a pharmaceutical company. Turns out there were some drug salesmen that needed some information to sell drugs STAT, so Ted headed for Newport while I headed to the post office to overnight a mic (the beloved Studio Projects C1) to Georgia.

The tapes (and mic) came to NY yesterday, and we spent some time listening to and loading the contents last night.

These kids can freaking sing, to my very untrained ears. The music is powerful, emotional, and inspirational (especially considering my atheistic slant); my favorite parts are those sung in Latin, or Italian, or French. (I think those are the languages. There may be some Spanish in there as well.) In English it's a bit too literal. I like my praise vague.

But what about all the technical minutiae, you say. Do not fear!

The recording quality is quite good, although it was done on blackface ADAT and has the limitations inherent in the medium (timecode dropouts are always an issue with those machines, and they're 16 bit). The C1's (mine and Pat's) worked well as a stereo pair. There was an AC running for most of the session - it was the old AC vs. Comatose Performers conundrum - which is audible on the quiet parts and masked on the loud parts.

And man, are there quiet and loud parts. A choir has an enormous dynamic range, already above the limit of a 16 bit medium (with a theoretical dynamic range of 96dB) and not far from the limit of 24 bit (ideally 144dB).

So there are two issues, intertwined, that we will deal with as we mix and master this project.

First is the dynamic range. Compressors and limiters are the usual tools for handling this, but in choir music (like classical) compression can really change the feel of the performance. It's not like rock and roll, which is built around compression (guitar amps compress, tubes compress, tape compresses, the radio compresses...); choirs are meant to be listened to live with all the dynamic range the ear can handle (120dB before the threshold of pain). But once you put them on tape (or disk) you have to deal with the noise floor of the medium.

We started by normalizing. While I normally don't like normalizing (it's more math, the bane of digital), in this case it makes sense. I need as much level as I can get going into whatever the next part of the chain will be or my noise floor will increase. Normalizing brings everything up until the loudest part is at maximum volume (what they call 0dB in digital - and if you find the whole dB thing confusing, you're not alone), keeping us from guessing how much gain we should be adding. George Vitray has advised that choir recordings are often not limited at all, and I'm inclined to go with him on this one. Listening to this recording should be like being there, not a combination of what we would hear and what we want to hear (the usual case for recording). Perhaps some very very slight limiting, only to eke out a few more dB (noise floor, everyone).

In terms of the AC, we'll use very minimal filtering. We started tweaking filters last night, and we'll likely end up automating them, using them as needed (ie on quiet parts) and shutting 'em down when things get loud. With a choir you have to watch your EQ'ing, as you can easily thin the sound and take voices right out. The human voice has a pretty wide frequency range. We'll be careful.

So once we've got the dynamic range and AC under control, it's a simple matter of separating the tracks into individual songs, putting them in order, and off to Georgia it goes.

I hope the kids like it. We'll keep you posted.