ONE-POINT OF VIEW

Ever since my visit to Denon's 1986 Gustav Mahler recording session in Frankfurt, Germany, I have been having second and third thoughts about stereo miking. There is an increasing new ferment in that area, indicating, perhaps, a slow reversal from the current and almost universal "one-point" or coincident stereo miking, which sacrifices the valuable musical perceptions of out-of-phase information between channels, in order to avoid a host of technical troubles that can plague us in just about any of the established stereo media. But there is technological change, some of it drastic, as we know. And the real values in the wider separation of stereo micro phones persist and are re membered. Yes--there are the experimenters, too, the inquiring minds who will not let such a good thing, any possible thing, go untried.

That's what I mean by a "ferment."

You may remember that Denon makes a big point of its "one-point" technique for Mahler's huge music, the main mike array abetted by "assistant" mikes re-phased to eliminate the factor of distance. As I observed this setup in a very large hall, I thought that I saw not a single compound mike but, rather, two units mounted perhaps a foot or more apart. I may have misjudged this, but in such a large hall, a foot-wide array could be called a "point," as a basket ball in a stadium. Nevertheless, even a foot of non-coincident separation could have an interesting impact on the sound. Through phones, it would rate as hyper-binaural, wider than the spread of two ears. Through loud speakers, it might add a lively touch of out-of-phase information. Hopefully, that was the intention.

As any audio man knows (and most consumers don't), there are two apparent types of perception in our stereo listening, volume and phase. The most basic of these is the perception of multiple phase differences, tiny differences in time of arrival at the ears, because this makes the other possible--the selective pinpointing of volume differences whereby we can sense directionality, right across our stereo "stage," from side to side. The volume proportions between each channel, selectively, element by element, are what does it. Switch to mono, and you can see how it works. The two speakers now radiate the same signal, no inner differences. You hear a central blob of sound, and if you slide your balance fader sidewise, that blob moves, too, from side to side and all points between. In coincident stereo, with two mikes (via various different arrays) essentially in the same place, but sharply aimed to each side in order to emphasize volume differences, we hear all those positions at once, and even the spread of liveness or room ambience which surrounds the music.

Why bother with recorded phase differences when the microphones are actually spaced apart so that arrival times at these two microphones are quite different?

Those multiple phase differences, if & when, do not really do much for specific directionality. It is the relative volume that does that job most exactly.

So why discuss phase at all? Well, very few of us do, in all truth.

Phasing is the well-known sleeping dog to let lie, if you follow me. Forget it.

Coincident by-the-volume stereo is remarkably good, and never more than now-nor can we forget that "one-point" has other important attributes going back to mono days. The theory even then was that in every hall or "venu," as they say, there is theoretically one ideal point, above and out in front of the (classical) music, that is optimal for balance between the musical instruments or voices and, even more important, between the direct sound to the mike and the surrounding reverberation or ambience. Indeed, once found, that point proverbially (and in truth) seems to jump out at you with astonishing realism. It was no old wives' tale but a real phenomenon (ask Bert Whyte!) which you may remember from the famous Mercury mono LP recordings.

Now, as Denon made very clear, that same effect is embodied in their "one-point" two-channel stereo mike array, and for the same good reason. (We all know about the contrary technique that swung away from the single mike, multi-mike mixdown. It had its values, too, and still has them, but that is another subject.) What, then, does the stereo separation of micro phones do for stereo listening? (Those tiny phase differences-tiny even though to a lot of audio equipment they are all too enormous.) Not specific directionality. Ambience.

That is the memorable result, and once heard it is not forgotten. An instantaneous, startlingly realistic room sound that is, in fact, essentially what you hear in a live situation. You can almost move your head quickly and follow the bouncing reverberations, this side, that. Not easy to describe in words! But it is, in its own unique way, a priceless asset in the reproduction of sound.

We do get a bit of that direct phase ambience in many current recordings, but only because they are not rigorously one-point. Even a little bit helps; the recording simply sounds better, more immediate and realistic. Indeed, it is a tantalizing, tempting thing! Every recording engineer wants his product to sound its best, after all.

With care, and good equipment all the way, you can indeed use a wide separation of stereo microphones, and, in fact, that sort of array was regularly employed in most early stereo and sporadically right on up through the years. You do risk trouble, in LP cut ting, in playback, in broadcasting and so on. Ingenuity, all the way, can get around a lot of it if the will is there--and the desire.

We must understand that this business of liveness, reverberation, and ambience has been a part of most mu sic for centuries and became very much a part of electrical recording around 1925, when sound engineers (and record producers as well) first began to realize that a "live" sound was musically nicer than the dead sound of acoustic recording. Minus electronics, the old recording horn could not hear ambience.

Performers were necessarily very close and very loud. Listen to any old disc of that sort, and you will see. Curiously, the early electrical recordings with microphone are still dead; it had not occurred to anybody to try any thing else. (Even the pictures show curtained and soundproofed studios, the normal rule of the time.) From the early recordings onward, the micro phone was given an ever more "live" sound as the values of ambience were better understood. Even so, the best classical recordings of the 1930s sound to us relatively dead. The attributes of hi-fi--sharper, clearer definition, lower noise and so on-made still more ambience desirable. Well before stereo, with the "one-point" mono technique, we had reached an early optimum. So you can see that liveness, spatial ambience, is not necessarily dependent on two-channel phase differences.

Nature has a habit of being redundant, in the computer sense as well as digital coding for audio. Natural perceptions are always compound, different aspects of "signal" reinforcing each other, different senses working strictly together, rather than separately. You can believe that directional hearing is immensely sharpened by directional sight! You see it, you hear it, it's there. But, luckily for modern entertainment electronics, we can get along well even when some of the natural signals are missing, like the 1934 16-cylinder Cadillac running on 15 cylinders. Indeed, we seem to thrive. As per my recent comments, we can go even further. We can wonderfully learn to take contradictions between our senses, between what we imagine and what we know, as in the portable con cert hall in your automobile. So we have, indeed, been able to appreciate the sound of musical ambience even without the benefit of stereo to enhance it. We really did wonders, and the realization that so much could be so well achieved without stereo (and hence without the technical phase problems) gave a great and much de served boost to most high-quality re cording. This is the background of coincident stereo.

What technical phase problems? Read the technical journals, please, if you are not familiar with them in your own pro audio experience. I give but two easily assimilated examples. On stereo LP, the out-of-phase information appears as vertical grooving. The mono or "same" elements between the two channels are lateral cut, just as in the old mono disc. There are all stages in between, of course, from zero difference to 180° opposite. A hefty, big bass note that is highly out of phase as recorded (from different points in space) can be lethal. Loud lateral (mono) grooves can be spread out to make room for violent sidewise stylus movements, but a whopping vertical cut, if it can be cut, has nowhere to go but up or down. Many a tortured stylus has thus bounced out of control, pushing the next groove at the bottom, hop ping into the air at the top. Big sounds out of phase are anathema to the working disc cutter, and to heck with that wonderful ambience effect.

Moreover, as we all ought to know, a stereo voice recorded out of phase tends simply to cancel out when the two signals are played mono. What, you want the stereo Michael ("I'm bad!") Jackson faded to nothing on a million mono players and radios just because he's out of phase? A fine excuse never to explore the beauties of variable stereo phasing. Even though the microphones aren't bothered a bit, nor most of our separate circuitry, nor your loudspeakers nor, most important, your ears! That's the psychology of it. In stereo, the coincident system offers the kind of rock-solid guarantee that big biz wants and needs.

The truth is that coincident stereo is actually multi-point mono. With all its virtues, it lacks the ultimate and best value of stereo sound, sacrificed as an all-out protection against problems that often can be solved.

I suppose that because radio has the most devious problems in respect to phasing between stereo channels and because it has had the law after it, too, for so many years-it is the most institutionalized medium in respect to coincident stereo. After so long, since the '60s, the original necessity to do something about the phase problem has solidified like a layer of geological sediment into the hard sandstone of Procedure. On Public Radio, it seems, coincident stereo is mandatory for all its operators, and there are even work seminars for the engineers to make certain the rules are followed. (More on this later.) Recording companies, I suspect, follow the coincident route on a somewhat more aesthetic ground; they are convinced that it is the right answer for optimum sound, problems or no.

And yet, there are always those questioning minds, the experimenters, the gadgetmen, who want to try a thing just because it is there. Many audio people know that phasing effects can be intriguing in stereo, so why not try, just for fun? Moreover, technical advances have made this a lot more interesting in two basic ways: First, ever greater ingenuity gets ever closer to eliminating or reducing the familiar problems--witness Shure's series of "trackability" test LPs, each one getting more demanding than its predecessor. Second, the gradual retirement of the LP and, likely, the audio cassette (in favor of the much superior DAT) the general advance of digital everywhere--goes a long way to deflate the problems' importance.

It's simple. We are closer, day by day, to being really able to use more phase contrast in our recording. And that inevitably means we will. Some body will, anyhow. Multiple mono is just not quite good enough.

(by: EDWARD TATNALL CANBY; adapted from Audio magazine, Jun. 1988)

= = = =