April / May 2009

Closer To The Absolute

  Reader Malcolm Balfour's letter in this issue takes me to task for comparing the sound of Reference Recordings' HRx 176.4kHz/24-bit files to the sound of live microphone feeds [Issue 190]. Mr. Balfour contends that I should have been comparing the sound of high-resolution digital audio not to microphone feeds, but to the absolute sound — the sound of acoustic instruments in a real acoustic space. He rightly points out that microphones are inherently colored devices and thus one step removed from the absolute sound.

Although I'm acutely aware of the limitations of microphones, I defend my statement. Different microphones will produce different interpretations of the actual acoustic event, but live microphone feeds — no matter which microphones you use — possess common characteristics that one simply doesn't hear from analog tape, standard-resolution digital audio, or LP. Those characteristics include a sense of life and transparency, of transient quickness without etch, of resolution of extremely low-level detail, and of a complete ease and grace. When that microphone feed is stored on analog tape or converted to CD-quality digital audio, much of the magic disappears. Low-level detail is scrubbed off; the sense of space shrinks; timbres aren't quite as pure and liquid; transients aren't as startling; and the entire presentation is a disappointing step down from the microphone feed, which is itself a disappointing step down from the original musical event.. But high-resolution digital audio done right (a rarity, by the way) maintains the vivid sense of immediacy that one hears from microphone feeds, even if those feeds inevitably deviate from the absolute sound.

Mr. Balfour also seems to suggest that I regard microphones as perfect transducers. Far from it. In fact, I once created a demonstration of the colorations inherent in even the world's best microphones. I assembled a collection of top microphones to record J. Gordon Holt reading one of his classic essays, switching microphones every few sentences and editing the essay together. The result appears on the first Stereophile Test CD. Listening to this track is startling and sobering.

Moreover, as the teacher of a college recording-engineering program, I consistently reminded students to keep in close contact with the sound of actual instruments. We would set up a pair of microphones on, for example, a piano and listen to the microphone feed in the control room. The sound invariably needed work, so I'd ask the students what they would do next. The students' answers were always along the lines of applying equalization or other signal processing, moving the microphones, or changing microphones. My reply: "Let's go back into the studio and listen to the actual piano and then choose a course of action."

I believe that the primary reason reproduced music doesn't sound like live music is some kind of catastrophic loss that occurs at the microphone diaphragm. Ed Meitner and I once discussed this subject at length after he told me about an experiment in which he put a microphone in front of a guitar amplifier's 10" cone and then tried to reproduce the sound through the same cone that created the sound in the first place. The reproduction sounded significantly different from the original. Meitner chose a guitar amplifier to keep the challenge as simple as possible; if we can't accurately reproduce the sound of a vibrating paper cone with the same cone that produced the sound — never mind an orchestra in a hall — we have a long way to go toward realizing the absolute sound.

The reference for judging the quality of music-reproduction devices will always be the sound of instruments in an acoustic space. Comparing high-resolution digital audio to a microphone feed isn't an abandonment of that ideal. Rather it's simply an observation that the degradations traditionally imposed by storage media have been reduced to the point that home listeners can now move one step up the chain — one step closer to the absolute sound.