Stereo in its most basic form is a robust medium. One can hear the ancient demos of ping-pong games and trains traveling from left to right or right to left on cheap computer speakers sitting on a glass tabletop. But if you want to hear the real sonic glories that stereo is capable of—the insight into the original recording venue, the sense of expansive space, the depth and width of image, the focus and location of instruments, and all the rest—then you need not only good equipment but also very careful setup of the system in the listening room. Everyone agrees on this in theory, I think, but in practice I have noticed that many audiophiles and even some reviewers do not follow as carefully as they could the basic principles that make things work for optimum results. So it occurred to me that it might be useful as a reminder to summarize what I think are the most important of these principles. Most of what follows will likely be at least roughly familiar. But the specifics will be of interest, I hope. And the last topic is something I think relatively few people have considered carefully. With that, here are the techniques to optimize the stereo experience.
1. Keep a clean path from speaker to listener
Stereo in the strong sense depends on subtleties and especially on subtleties about the direct arrival of sounds and the sounds that arrive very soon after that. Later sounds play a role in what you are hearing overall but they are less crucial to stereo as such because of the well-known “precedence effect” (later sounds are treated as unimportant by the ear/brain in deciding on spatial and locational information).
Now what do I mean by a clean path? You need to have nothing whatever on the floor or anywhere else close to the path from the speakers to you. The floor (and ceiling) need to be smooth. The floor should be smooth (think carpet). High-frequency reflections are thus minimized. If the ceiling is not really high, it would be nice if it could be soft, too, but at least it can be smooth. The point here is that rough objects—equipment items, furniture, and so on—scatter sound irregularly. This is a serious cause of the blurring of one of the major things that makes stereo work ideally—namely, timing cues in the higher frequencies.
Some years ago, Demian Martin did a blind demonstration (with me as the subject). He had me close my eyes while in the listening chair, and to tell him if I heard a change in the sound, and what that change was. There was no doubt when the change happened. The stereo went to pieces—it became defocused, and the sense of space was seriously damaged. After I said this, he asked me to open my eyes. What he had done—silently—was to lay a record cover on the floor symmetrically between the speakers and a little in front of them, not even on the path from the speakers to me, but fairly close to it. This was not much of a change physically, but it was truly destructive of the stereo performance.
What price then in stereo terms does a coffee table in front of you, between you and the speakers, exact? And yet one sees such things often enough. Of course, it is convenient to have a table in front of you for the remote controls, for a drink, or a few discs. But it really won’t do. And anyone who is listening like that is not really hearing what stereo is capable of.
I recall also Kevin Voecks describing to me a system where the owner had covered the entire floor with diffusers—not an inexpensive thing to do. Voecks remarked that this showed real dedication. Then he added ruefully, “But of course there was no stereo imaging to speak of.”
2. Clear the decks next to you
Along the same lines, hard objects close to your head are bad. The best thing to sit on is a stool or a chair with a very low back, so nothing or nothing much is up near your ears. If your back won’t take this, at least use a chair that does not come up to head level. And make sure it is soft. Hard leather sofas are the pits—they reflect high frequencies. If you must sit on one, cover the area behind and near your head with a blanket. What you do not want is a lot of very early scattered high frequencies coming at you. But the low-back chair or stool is best. Try it. You will be surprised by how much better your system sounds.
3. Symmetry rules
Stereo is based on symmetry between the two channels. For late arrivals, where one is encountering what amounts to a diffuse field in the treble, the role of symmetry is reduced. But for anything like early arrivals, and in the lower frequencies for the sound all together, symmetry is indispensable. In the old days, when a lot of audio people did not understand how stereo actually worked, and in particular did not fully appreciate the importance of interaural time differences, many would try to adjust the differences between the two channels by electronic means—balance controls and individual equalization of the channels. The result, of course, was a disaster, because analog EQ and balance controls could not deal with the time asymmetries in an effective way.
As it happens, sufficiently detailed DSP can, in fact, correct these problems. But it is much better to have the whole setup as symmetrical as possible from the start. As always, DSP room correction works best when there are fewer and less severe frequency-response peaks and dips to correct. (The one exception is very broadband bass corrections, which are useful in practice.)
Part of the symmetry business is you, the listener, being equidistant from the speakers. In general terms, people usually do this. But they do not always do it carefully enough.
Let us consider the numbers. People can hear time differentials between the two ears of about 1/100,000 second. Let us be conservative and call it 1/50,000 second. How far does sound go in 1/50,000 second? Well, one millisecond (1/1000 second) is about a foot of travel. So 1/50,000 is about 12/50,000 feet. That is, about a quarter of an inch. So if you want to get what you can out of stereo speakers you need to match your distance to them within a quarter of an inch. Time for a tape measure! Mark the spot and stay there.
4. Sit close to your speakers
There are always going to be some early arrivals that one really does not want. One tries to make them small in amplitude and symmetric, but there are bound to be some. The way to minimize their effect is to make the direct arrival as large a part of the early arrivals as you can. Sitting close is how you accomplish this. Of course, you cannot get too close—most speakers depend on a certain distance for the drivers to integrate correctly. And to some extent one needs some distance to avoid hearing micro-vibrations that might make the speakers more apparent as sources. But you will get the best stereo if you sit as close as reasonably possible relative to these points. Many people sit much too far away and end up involving the room far too much. Sitting closer to the speakers changes the ratio of direct-to-reflected sound.
You should also be far from the wall behind you, and the speakers should be far from the wall behind them (unless they are in-walls or designed to be placed against the wall). Indirect early arrivals are not good, and listeners and speakers positioned near walls are also detrimental to good imaging. Fortunately, sidewall reflections, even early ones, are not nearly as bad, though getting rid of them is good, too.
5. Have a soft room
Recordings are made with the expectation that there will be some room sound added during playback. Few recordings are made with the expectation that they will be listened to anechoically. But there is effectively no chance at all that you will get too little room sound. This is very unlikely in a domestic environment.
In my view, the best stereo arises in what is known as an RFZ (reflection-free zone) room. In such a room, the early direct arrival is all there is for quite some time, and early reflections are absorbed. After you hear the direct sound, the room sound appears as a largely diffuse supporting sound in which discrete reflections are not in evidence and/or are arriving too late to be heard as part of the early sound. The result is that the direct sound is heard distinctly from the room sound rather than being confused with it.
Building an actual RFZ room is challenging, but one can get a good deal of the effect by having an acoustically soft room, particularly in the treble which has the largest effect on this phenomenon. Absorbing the treble component of reflections is a good thing. A soft acoustic environment is also more pleasant tonally, but it is the stereo effect that I am really concerned about here.
Glass walls are the worst. If you must have picture windows, get some heavy drapes you can pull over them while you listen. Years ago, Peter Lyngdorf brought to my apartment in Copenhagen a system including room correction. This apartment had a spectacular view—it was on a high floor and had one long wall of nothing but glass with steel supports spaced along them. But Peter remarked, and of course I agreed, that the room was hopeless for sound unless I could add some curtains. (I was not staying long, so I just lived on live music for the time I was in Denmark. Audio was a lost cause).
6. Sit still, sit really still, and close your eyes
Part of being symmetric is to stay in one place. Unless you move straight backwards or forwards, you will move into an asymmetric position! But there is another part of sitting really still that is more subtle but has important effects. To get into this, we need a short excursion into how perception works.
This is a complex subject so I can only give an outline. But roughly speaking, when the brain constructs a perception from raw sense data a lot of processing goes on. What we “see” or “hear” is not the same as the light in the eyes or the sound in the ears. It is a mental construct from that. Now this mental construct typically includes the following surprising thing: The construct uses many cues, but if one cue is definitely missing, the brain notes that it is missing and simply reprograms itself to use the other cues only.
This perhaps sounds like psychobabble, but there is a simple visual experiment you can do that illustrates the principle. Look at a photo with a lot of three-dimensional stuff in it or at a painting with strong perspective. Looking at it will give some impression of three-dimensionality. But now look at it with one eye closed and the other eye looking through a short tube that blocks out the background and lets you see only the picture.
There is a startling increase in perceived three-dimensionality. If you have never tried this, you will find it hard to believe your own eyes—or eye.
What happened? When you looked with both eyes and with the background around the picture visible (the way one would ordinarily look at a picture), the brain got some cues of three-dimensionality from the perspective. But the brain also received information that the surface on which the picture was painted or printed was in fact two-dimensional. Binocular vision showed it to be flat. And the background also showed it was flat. So the brain was getting contradictory cues—and it let you see a little three-dimensionality but not all that much.
When you look with one eye and no background visible, the cues to the image being flat are blocked. With no binocular vision and no background—with some cues that one is looking at something flat gone—the brain goes with what it has left: the photo or painting’s actual image, and that says 3-D in a big way!
(This effect is well known—I definitely did not discover it. It is discussed for example in Francis Crick’s book on consciousness The Astonishing Hypothesis, which is where I first read about it. Ever since, I have gone around art galleries with a short tube of one-inch diameter!)
Now back in the audio world: One of the ways that the brain detects the location of objects and the shapes of rooms, of space in general in auditory terms, is by moving the head slightly and noting the changes in the sound. This is basic, for instance, to telling front from back (interestingly this was discussed in the Ask Marilyn column in Parade magazine not long ago). In particular, head movements are providing, for instance, subliminal cues as to where your speakers are and what the acoustics of your listening room are.
But as far as I can tell—and I admit I am going on experience here and have no scientific documentation, though there may be some—when one holds one’s head exactly still while listening to stereo the brain is no longer thinking about the head-movement cues. Head-movement cues are missing in stereo as such. For virtual reality one has to do something about that (hence the head-angle sensors for VR headphones). When you cancel those cues by simply not providing them either for the stereo signal or the room around you, the brain locks in more on the other cues being provided, specifically the cues provided by the stereo signals themselves.
This has a substantial effect in my experience. One has a sense of suddenly “sinking into” the stereo space. I cannot guarantee you this will work for you, but it is worth a try. And one gets used to sitting precisely still. This tends to put one in a sort of meditative state, which is in itself not inappropriate to listening to music. And, to my ears, the stereo gets better.
Finally, closing one’s eyes also removes cues that one is not in a concert hall or a stadium venue but in your listening room listening to speakers. I won’t say much more about this, since I suppose everyone has tried it. But do give the sitting absolutely still a try if you have not.
So that is it: Sit with eyes closed, head absolutely still, on a stool close to the speakers in a soft room and in a symmetric position exactly equidistant from your speakers with speakers far from front walls and yourself far from backwalls, and nothing on the floor between you and the speakers. If you are not doing all these things already, I can all but guarantee this will be a revelation of how well stereo can work. Stereo can seem like a miracle if you do it right. And to cap it off, all these tips are free.
Happy stereo listening!