In January of 1998 I was contacted by the Coolidge String Quartet about making a recording for the Danish record label Olufsen Records. The recording was of particular interest to me because it included three works that are rarely performed, let alone recorded. The first was Benjamin Britten's first string quartet; Britten was the most familiar of the three composers. The piece was written while the composer was in his 20s, but don't let his age fool you; this work displays a prodigious understanding of string writing and a mastery of theme and development. Danish composer Anders Koppel was next on the schedule, with his "String Quartet Number 1," and the third was a 1990 composition by Greg A. Steinke entitled Native American Notes: The Bitter Roots of Peace. Each piece had a neoromantic flavor, which meant miking all three similarly, though some subtle changes were necessary.
Anyone familiar with Britten's work will know that, stylistically, he has one foot in the past and the other in the mid- to late 20th century. Along with lush romantic harmonies and phrasings are Bartok-esque pizzicato and very forward-looking dissonances. To capture this on tape, our plan was to combine a classic minimalist approach of two spaced microphones relatively close to the ensemble, and a pair of omnis in the room for ambient pickup. The two other pieces worked well in this configuration but needed a little more "edge" to fit with the more contemporary style of composition. For the Koppel work, simply bringing the ambient mics down in the mix by about 10 dB and moving the mains toward the ensemble a couple of inches worked great. The Steinke, however, required a more direct sound, which meant moving the mains down almost a foot closer to the ensemble, and the ambient mics were lowered to the point of almost being out of the mix altogether. The Steinke also included excerpts of Native American poetry interwoven between short musical interludes; the voice-over was recorded weeks later in the studio directly to Sonic Solutions in order to expedite the editing process, which had already begun.
CHOOSING A PLACE TO RECORDClassical recording requires a performance space that will become an integral component of the sound canvas. This is not to say that other genres don't benefit from their acoustic surroundings, but the extreme dynamic range associated with classical performance makes it nearly impossible to work in a noncomplementary ambient space. Inherent characteristics of a remote recording space must include an extremely low noise floor, appropriate first reflection and reverberation, a separate small room with minimum reflections for a temporary control room and, of course, a use fee commensurate with your budget. For this recording, the Coolidge String Quartet and I spent several months visiting churches and concert halls all over Connecticut until we stumbled upon the Congregational Church of Naugatuck. The church's music director, Scott Lamlein, was extremely helpful in accommodating our needs, which included converting the minister's study into a control room, closing the sanctuary for a full week (not easy to do in a church) and cycling the heat to down times.
The distance from the altar to the narthax measured just over 50 feet with the width only slightly smaller. An ornate ceiling arched over the pews, with its apex running front to back. On the sides there were five Corinthian-style columns, each about five feet from the wall. Along with these physical amenities were the usual detriments often associated with location recording, including a nearby commuter train, adjacent roads and poor electrical ground. We settled on this location, though, because of the lush tonal color produced by the strings, and we just took a break when the ambient noises became prohibitive.
THE MICROPHONE SETUPI've always subscribed to a minimalist approach when choosing a microphone configuration. When more than two to four microphones are used, the result is often a collapse in the sound field's depth. We scheduled time in the church a month before the recording sessions to experiment with different microphones and confirm that the space would be appropriate. Bruel & Kjaer 4011 cardioids spaced slightly smaller than the width of the ensemble were used for the main pair. Their distance from the ensemble sat just outside the reverberation radius for the Britten, and moved within the direct sound field for the other pieces. (The "reverberation radius" is the invisible line that separates the area of direct sound from the diffuse. It's important to note, however, the reverberation radius as it relates to the polar pattern of the microphone: An omnidirectional microphone will sense this balance close to where our ears will, as our ears exhibit essentially an omnidirectional pickup. Cardioids, on the other hand, are by definition more sensitive on axis and therefore must be brought back farther-1.7 times as far to be exact-to achieve that same balance. Logically, the more directional the microphone becomes, the farther back it must be placed to achieve that equilibrium. It's not always necessary to place microphones right at this point, but it's a great place to start, and you know exactly what the result is going to be if you move them forward or back.)
The B&Ks in this configuration provided a wonderful sense of space, a wide stereo image and a brilliant roundness to the sound. The Britten and Koppel pieces needed just slightly more reverb than recording with the two mics provided, so two AKG 414s in omni were placed about 30 feet back, with 20 feet of separation, and about 15 feet high. The 414s are inherently a bit weak in the high end, so they were exactly what the doctor ordered for reverb color.
THE CONTROL ROOMThe minister's study was used for a control room and needed just a little temporary renovation. Absorption was placed on the sides of the listening area to defeat some of the distracting high-frequency reflections produced by the near-field speakers, and behind us to break down rear reflections as much as possible. A room mode around 400 Hz was difficult to defeat, but switching between loudspeakers and headphones gave us a good idea of what was really going to tape. The B&Ks were sent through a brand-new Earthworks LAB102 microphone preamp on loan from Parsons Audio (Wellesley, Mass.), and the 414s went directly into an Audio Developments AD145 Pico mixer. A Mytek 20-bit A/D converter followed the mixer, and the recording was made on an Otari DTR-8 DAT machine. Mytek's Hi-Bit 16 was used for dithering from 20 to 16 bits. Monitoring was mainly on Genelec 1031As with occasional reference on a pair of Beyerdynamic DT770s (mostly to listen more closely to unwanted ambient noise). The digital output of the DAT was sent through the Mytek's D/As for better resolution monitoring.
As is common in many older New England buildings, the integrity of power was of major concern. The usual culprits such as RF noise and fluctuating voltage were there but easily tamed with an isolation transformer. The biggest electrical concern was the lack of a quality earth ground. Although we weren't able to rectify the situation entirely, some relief came from grounding the transformer and the Otari DAT to an old steam radiator in the control room. The situation was further complicated by a few days of dry weather resulting in frequent static shocks from the mixer and anything else that was touched, usually at the most inopportune moment, of course. A couple of times these shocks were of enough intensity to throw the Mytek into a tailspin, only to be restored by cycling its power.
EDL: DIGITAL EDITINGPlanning the post-production process while making a recording is a key to overall efficiency. Creating a rough draft edit decision list (or EDL) during the recording sessions can help expedite the process later on. Producer Eric Dahlin worked from an 11x17 sketch pad that had two systems of music pasted on each page. This system is the best I've worked with for providing ample room to make editorial comments and other post-production notes. You should always keep in mind that it is impossible to make too many notes in the production process. Even the most experienced producers will not remember all of the subtleties within each take or insert.
Back in the studio at Metcalfe Productions, all takes were loaded onto a Sonic Solutions system. For projects of this magnitude I am more inclined to load in all of the takes rather than only those selected in the producer's decision process. Inevitably, when the clients come in to listen to the edits, they will find more problems that need to be addressed, and if all of the music is right there in the computer, the searching time can be drastically reduced compared to digging through the session tapes. This also supports the argument for recording directly to the computer and using the tape simply as a backup source. With PCI expansion chassis now being developed for Mac-based laptops, this is becoming less of a chore than pulling a large system out of your studio.