As We See It

I had an experience at last summer's CES in Chicago that bordered on the religious. I heard the legendary $42,000 Wilson WAMM system.

Apart from the fact that designer Dave Wilson chose not to play any symphonic or operatic music, explaining that "there aren't any really good recordings of that sort"—note that, Keith Johnson!—the reproduction was by far the most realistic I have ever heard. Wilson was playing copies of some of his analog master tapes, the most sonically impressive of which was a short segment of his wife Sheryl-Lee's soprano voice with pipe-organ accompaniment, done in a large acoustical space that sounded like a church.

I was struck by four things. First, the tonality of the sound—its vowel quality, if you will—sounded exactly and precisely right. So right, in fact, that it was hard to believe this was an electromechanical reproduction I was hearing. Second, the frequency extremes seemed limitless, as though the system could effortlessly reproduce anything from a bat's radar to an earthquake, yet each extreme was so well-controlled and in the right proportion that it never called attention to itself, even when the pipe organ was rattling every object in the room. Third, the stereo imaging, depth and spaciousness were more convincingly reproduced than I had thought possible. The loudspeakers were simply not there. The sound was above, below, between and beyond them, as if the back wall had been removed and the room opened out onto the performing space. And fourth, as a direct result of this, I was made aware of a major failing in all perfectionist audio systems: I was not in that performing space.

In front of me was the sound of the church in which the recording had been made. To the sides and behind me was a a fairly large hotel room. As convincingly real as that frontal presentation was—probably because of the realism, in fact—I found the falseness of the rest of my surroundings quite disturbing.

The next day I heard the Infinity IRS system (the Mark III version), at Infinity's much larger demo room adjoining at the McCormick Inn. Arnie Nudell and Bascom King were in attendance, and their signal sources were taped copies of some Reference Recordings masters, including Dafos. This was the third time I had heard the IRS—each time at a CES—but it was the first time I have been really impressed with what I heard. I would not hazard a comparison between these two supersystems on the basis of such short listens and such disparate program sources, but the IRS also amazed me with its imaging and its reproduction of depth and spaciousness. In fact, when I heard one section of Dafos that opens with bird calls—a section I have heard before—I was immediately and for the first time aware that it had been recorded indoors. Obviously in a large space, but indoors nonetheless. In fact, it had been done in the rain forest exhibit at a zoo.

Hearing these cost-no-object systems crystallized an idea that has been rolling around in my semiconscious for some years, and forced me to a conclusion that is going to strike a lot of Stereophile readers as heresy:

Perfectionist audio needs surround sound!

That audiophiles value spatial information in recordings is common knowledge. The warmth and richness and breadth of well-recorded acoustics in a fine hail add considerable excitement to the sound of large performing groups. And record companies have been seeking to deliver this quality in their recordings. As an illustration of what I'm talking about, compare any recent recording of, say, the New York Philharmonic with one made during the late 1950s or early '60s. The early ones sound inordinately dry and lacking in spaciousness. Modern recordings have just about the right amount of space surrounding them, but it's all in the wrong place!

In a live-concert situation, the hail reflections that we hear as spaciousness come mainly from the side and rear walls of the hall. The sounds coming directly from the instruments in front of us are unreverberated and, save for a certain amount of high-end attenuation due to air attrition and a small amount of short-interval reverb from the rear of the stage shell, are essentially the "raw" sounds of the instruments. The front-only sound is in fact quite like what we hear from those pre-1970 stereo recordings—rather dry and somewhat overly detailed. It needs to be mixed with the reflections from wall surfaces to provide the richness and blending we hear at a live concert. The effect, when we're in the hall, is detail in front of us and space all around us.

We can approximate these qualities by hyping the spaciousness that we record along with the front (stereo) channels, but we are no longer surrounded by space. The hall is in front, the listening room is around. If we add a few extra loudspeakers around us, and a device to feed spatial information to them, we get a more natural spatial presentation but we destroy the detail from the front channels. Why? Because the instrumental sounds are already "blended" adequately by the front-channel ambience, and any more blending from rear channels adds too much.

I have become convinced that if "concert-hall realism" is truly what audio perfectionists want, we are going to have to start thinking seriously about surround sound. It simply won't be achieved otherwise. Those front two channels are just getting too revealing of spatial information for us to continue to pretend that it all belongs up front. We need at least two more channels to give side and rear ambience, and while the Compact Disc is claimed to be able to accommodate two more discrete signals without loss of playing time, the analog disc cannot. For analog, two more channels means matrixing. And what does that mean to most of us? Dreaded SQ!

Maybe you don't remember SQ. It was CBS's answer to the "fading novelty" of stereo. And it was a great idea in theory The late Benjamin Bauer, former head of CBS Labs and a classical-music lover, developed SQ for just the kind of ambience reproduction I'm talking about. CBS big-heartedly licensed the new surround system to allcomers, in the hope of having it adopted by every record company in the world. But then the marketing departments grabbed hold of SQ and all hell broke loose.

Convinced that the public (footnote 1) wouldn't perceive spaciousness if Agoraphobia were King, they mandated that orchestral recordings be done "in the round," with instruments blaring at the listener from every direction of the compass. As if that weren't enough, the SQ encoding was often badly done, most of the playback decoders were awful, and most SQ reproduction had so much leakage into the wrong channels that any semblance of proper imaging or sonic integration was lost. Perfectionists shunned SQ like a herpes lesion.

Actually it wasn't the lousy sound that killed SQ; it was the bewildering proliferation of competing and mutually incompatible four-channel systems, plus the public's realization that it would rather buy a new TV than two more audio channels. With both the perfectionist and JQP down on four-channel, it had nowhere to go but phooey.

I'm not implying that two-channel audio is as good as it can be or should be. And the Wilson WAMM and Infinity IRS systems are beyond most budgets. But affordable systems are already reproducing spaciousness well enough in front to make its absence in other directions conspicuously absent. The time has clearly come when we need some sort of spatial information that will put us in the performing area rather than on the outside listening in.

I mentiond my earthshaking obervation to Stereophile's Anthony H. Cordesman at lunch in Chicago, and I apologize to him for spoiling his appetite. But his reaction—that surround-sound systems are terrible and there's no point in adding surround unless it can be done right—did not satisfy me. After all, our inability to get the two front channels right for over 100 years was no reason not to keep trying. In fact, even now four-channel sound does not have to be as bad as it has been. The Tate SQ decoder and some demonstrations of the English Ambisonics system have shown that four information channels can be successfully encoded, and decoded, with none of the shortcomings of conventional SQ.

The problem with these systems, of course, is that they introduce a lot of extra circuitry which stands a very good chance of fouling the pristine front-channel signals we have worked all these years to purify. But additional active circuitry in a system need not foul the sound. Dynaco's tube amplifiers utilized the fewest possible amplifying stages in order to minimize signal pollution, yet Audio Research's amplifiers, with several more stages per channel, sound far cleaner than any of Dynaco's designs.

It isn't the complexity it's the sophistication of the system that determines its sound. There is no inherent reason why surround-sound decoders (and encoders, I might add) should be any more sonically polluting than an Audio Research SP-10 preamp (footnote 2). SQ encoding, for example, can be done using passive components. Active circuitry is needed only for input and output interfacing, and for getting back some of the signal lost in the matrixing network. But no-one has ever demanded that the active circuitry be up to audio perfectionist standards because designers know that audio perfectionists wouldn't buy the product anyway. Tate Audio is the first firm to strive for this level of quality, but no audio purist would allow such a device into the house. A signal processor? Yeccch!

But then audio purists are a slightly weird form of homo sapien—more emotive than sapient, more swayed by the gut than the cranium. They can rhapsodize over reproduced spaciousness and at the same time argue that it is enough to have that spaciousness convey only a quarter of the ambient information that was there to begin with. But most of them are also intelligent—far more so than your average bear. It is only a matter Of time before they become increasingly conscious, and then acutely aware, of the grotesque disparity between the spacious hall in front and the cramped living room elsewhere.

God knows, it took long enough for this to get through to me, and I'm not that much of a purist.

When a "consumer demand" for surround sound starts to develop, we'll see some efforts made to meet that demand. Perhaps then can be experienced what we mean when we talk about "immersed in the music."

---

Footnote 1: In the eyes of Big Business, the public is a nameless, faceless mother lode of inestimable wealth, totally devoid of judgment, sensitivity or taste.—J. Gordon Holt

Footnote 1: Don't count on it being inexpensive, though, or even as high a quality as Audio Research; others have made very expensive preamps that don't sound as good as Bill Johnson's designs, and I'm not convinced that a fancy decoder would introduce as little degradation as an SP-10.—Larry Archibald