Against the Dying of the Light: the Second Cantus CD John Atkinson on the Sound

Part 2: John Atkinson on Capturing the Sound of Cantus

The Chapel of the Good Shepherd on the campus of Shattuck St. Mary's School in Faribault, Minnesota—Marlon Brando and Sylvester Stallone are alumni—was built in the English Gothic style in 1871. The interior features collegiate seating and a high wooden "A" ceiling, and the floor is of stone flags laid over a suspended wooden platform. The acoustic is quite reverberant for a fairly small space, but on the bright side overall, and Cantus in full song verged on overloading the acoustic.

Placing the chorus on the altar behind the stub walls wasn't a good idea, so they stood in an arc in front of those stub walls, about a third of the way down the nave from the altar. I would have liked the singers to stand on risers, to minimize early reflections of their sound, but we had none, and so unscrewed and removed several rows of pews. Nevertheless, as you can see from the Sonic Solutions screenshot, which shows the waveform of the sound of a slapstick as picked up by the four mikes, there were still plenty of reflections.

As in my 2001 Cantus recording, I used two pairs of microphones for the main pickup: a spaced pair of B&K (now DPA) 4006 omnis, and an almost coincident pair of B&K (now DPA) 4011 cardioids. I used the ORTF technique for the cardioids: mike capsules 7" apart and angled at 115 degrees. The cardioids were mounted on an Audio Engineering Associates stereo bar fitted to a 12' high Manley Starbird stand. For one evening session, I also used a pair of DPA 4060 miniature omnis, one mounted 4" out from either side of a Jecklin disc for a parallel quasi-binaural recording. (My thanks to Len Moskowitz of Core Sound in Teaneck, New Jersey, for the loan of the tiny DPAs.)

For the Monday-night session I used a pair of almost-coincident crossed Neumann TLM103 1" cardioids to close-mike a Steinway piano for the Debussy Invocation. Before you inhale a gasp of shocked breath at the words "close" and "mike" in such intimate embrace, we had rented the piano for only this session, and had very little time for setup. I wanted to leave the main pairs of mikes optimally set up to capture the voices; only later, in post-production, would I mix in some of the close piano signal to counteract the rather distant piano balance that would otherwise have resulted.

Having set up the mikes in approximately the right positions—I try to match the same direct/reverb ratio for each pair of mikes, which mandates moving the directional cardioids farther back than the omnis—I set up my gear in the vestry, a side room off the main entrance, which necessitated running 100' cables from the six mikes. (You can never have too many cables!~)

All the sessions were done at 24-bit resolution and 88.2kHz sample rate. (The downconversion from 88.2kHz to the CD's 44.1kHz is demonstrably transparent, and the 88.2kHz masters can also be used for eventual SACD or DVD-Audio release.) Apart from the separate binaural recording, for which I used my Panasonic DAT recorder's A/D converters, all three ADCs' wordclocks were linked. This way the heterogeneous storage media would not be an issue once I'd uploaded all six tracks of data onto my Sonic Solutions Digital Audio Workstation's hard drives and aligned each pair in time with the others. (Each take lasted exactly the same number of samples in each medium.)