The Rhapsody Project Notes on the Sound part 3

It is very important when you're using spaced microphones to ensure that their outputs are in time synchronization. In other words, for a musician at the center of the stage, the sound from the instrument should reach all the microphones at exactly the same time. In the analog days, you couldn't do that unless you very carefully measured the distances from each microphone to the instrument and made sure they were the same. But then the microphones might not be in the right place to get the optimal sound quality from each. So the cardioid pair were placed to give the optimal sound picture, the omni pair to give the optimal tonal color. With digital recording and editing, the four channels could be synchronized later in post-production.

Each reel of tape started with the sound of Wes Phillips standing at the dead center of the stage banging slapsticks. This is exactly analogous to the "clapper board" used in movie-making, and gives a very nice, clean impulse on each of the four tracks of the digital recorder. Before the editing and mixing were begun, those four channels were uploaded into a Sonic Solutions hard-disk editing system running on a Macintosh computer. Each channel could be independently slid backward and forward in time until the sound of the wood blocks was synchronized on all four tracks. Anything that was in the center of the stage will therefore produce a signal at exactly the same time—or at least within one sample period—in all four microphone channels, and therefore, by definition, be in the exact dead center of the reproduced soundstage.

The ORTF technique was developed in France and gives a nicely defined soundstage from a pair of cardioid microphones, but the tonal balance lacks low-frequency bloom. The spaced omnis, on the other hand, give a wonderful sense of bloom and very accurate tonal color, but have mediocre stereo imaging.

In post-production, the cardioids were used full-range to preserve their excellent imaging. The treble was cut on the omnis, then the two pairs of mike signals were mixed together. The cardioids are the primary microphones in the upper midrange and treble, while the omnis give the sense of space and envelopment in the low frequencies. The result is, we hope, the best of both worlds: You get an accurately defined image, where you can almost look into the soundstage to see where the musicians are. But at the same time you get some of the sense of bloom that you would have heard had you been at the live event.

By using the minimum of miking, we tried to achieve a very natural, realistic soundstage with this recording. There were no spot microphones on any of the instruments, which helps preserve the sense of depth. However, if you think about this, it does mean that instruments that were farther away from the microphones do sound farther away. Robert Dorer's trumpet in the second Prelude, for example, was at the rear of the orchestra and sounds distinctly more reverberant than, for example, the bass clarinet and the violins, which were at the front of the stage. To stay with this rather wide-angle perspective was a deliberate decision, however.

All the editing of the performance tapes was done with 24-bit resolution to preserve as much of the original quality as possible. But once the master edit list for the CD had been assembled, the problem remained: How to reduce that 24-bit data to the 16 bits mandated by the Compact Disc Standard? Simply dumping the output of the computer hard disks to DAT or CD-R, thereby truncating each digital word from 20 to 16 bits, both reduced the sense of recorded space and added a feeling of "digititis." I therefore "redithered" the data when I prepared the master using a Meridian 518 Mastering Converter, which manipulates digital data in real time. The input and output data word lengths can be independently selected; pre-emphasis or gain can be added if the operator so wishes; and, most important, the 518 applies a choice of noise-shaping curves to the music data. By shifting quantizing noise up to the inaudible 20kHz region as it reduces the master's 24-bit word length to the CD's 16, the 518 preserves as much as possible of the original's resolution in the midrange.

The maximum level on this CD was adjusted in the mastering to reach almost 0dBFS, the maximum possible on a CD. For those listeners with sound-pressure-level meters, 0dBFS on this recording corresponds to a peak spl at the microphone position of 110dB. If you set your volume control so that the opening of the first Prelude generates a peak level of 98dB at your listening position—which is loud—the playback level will be pretty much the same as that in the church where the recording was made.

I feel the combination of recording venue, microphone technique, and recording hardware makes Rhapsody Stereophile's finest-sounding CD yet; Hyperion's blistering performances are allied to accurately captured tonal colors and a recorded soundstage that is almost holographic. And Joe Cea's arrangement, in my opinion, breathes new life into Rhapsody in Blue—check out that delicious-sounding marimba as well as Tia Perdomo's creamy clarinet and Debra Taylor's bluesy 'bone! Buy a copy and let us know what you think of our efforts.

Enjoy!—John Atkinson