THE QUEST FOR AUDIO NIRVANA (3/3)
WHAT PRICE GLORY: PART III
This final 3rd installment of this series will undoubtedly cause mucho controversy. I’m sure that lots of nerves and egos will get rattled however, the truth is inviolate. True scientific enlightenment shouldn’t be contaminated by whims and a lot of marketing nonsense. It seems, unfortunately, that our industry has incredible amounts of both. So, without further adieu, let the war begin.
I believe that over the years, this entire industry has lost sight of what the goals of audio should be and that is the faithful REproduction of the live musical event. While quality aspects are almost as equally important, the primary function is still to create reality. If you went to a club and the piano had some of the unisons a little out of kilter, does this totally nullify the performance? I have heard Oscar Peterson play on a piano that had a few notes out of kilter but it did not diminish the performance. It was still REAL. The problems of reproducing the live event (or the studio version of the same) are truly monumental. As I’ve mentioned in previous writings, there are several pitfalls that make things painfully obvious that we have a long, long way to go. To paraphrase Peter Aczel of the Audio Critic a few issues back, there are many evils standing in the way of true reproduction. The first is the recording itself. What you get is what you get and you can’t make it better. You can make it differentbut not better. All of the information that you are going to have is already there, bad, good or indifferent. The next item is of course, the room and its acoustic properties. The third item is the loudspeaker. And finally, the lousy stereo format, which we’ve unfortunately been living with for the last 45 years. All of you MUST know that our electronics are the strongest link in this chain and not anywhere as weak as the other items.
I tackled electronics in part I and recording in part II. Now it’s time for loudspeakers and acoustics. It’s time to repeat for the umpteenth time the fact the ALL SOUNDS IN NATURE ARE CREATED IN MONO. I hope by this time everyone has had a chance to think about this and begin to comprehend this fact. Therefore and obviously, stereo reproduction is not real and does NOT occur in the real natural world. Just like digital, stereo is a concoction of mankind. Obviously the real question becomes “how can we make it better” or to be more precise, how can we make this more real. Easier said than done.
Enter the dragons
There has been for many decades, a general perception (incorrect, I might add) that a perfect loudspeaker would emulate from a point source. Nothing could be further from the truth. The key word here is “point”. Two “points” cannot convey an image with a total lateral as well as vertical dimension. Actually, I should say that NO vertical dimension can be conveyed and the lateral dimension is limited by the aperture. I know that a lot of you will want to shoot me for these statements but bear with me. Some of you may remember that many decades ago there were those that advocated a pulsating sphere as a perfect reproducer. Well, this may be true IF AND ONLY IF one is just trying to create sonic energy. And, it would also be necessary for one to be INSIDE the sphere. After all, many have thought if sound in nature is really spheroid in nature, then why not have a reproducer that creates spherical sound waves. Unfortunately, this just doesn’t cut it because sound sources DO NOT radiate acoustic energy absolutely equally in 360 degrees of arc. In addition, we can promulgate all the theories we want but the truth is that there cannot and never will be a “perfect” reproducer. It is not to be in this universe. The better question would be: how close can we really come? In order to begin to answer that question we must get physical.
There are 3 basic mechanisms of propagating sound waves that begin to satisfy the PRIMARY goals of stereo reproduction. These are one, a vertical slot at least 6 feet long and preferably 7 fThe second is a scalar triangular diaphragm with the top aperture at least 6 feet off the floor. And thirdly, concentric radiating rings ala the Quad ESL63 and its newer brethren. The vertical slot approach is very difficult, if not impossible, to truly implement properly. What is required is a slot width that is no wider than half a wavelength at the highest frequency to be reproduced, say 20 kHz. For this frequency it would require a slot width of about ¼ of an inch. Obviously, in order for this arrangement to reproduce frequencies down to around 200 Hz, it would have to have the world’s most gigantic magnetic structure behind it and be fed with the power of Niagara Falls. Obviously, there have been many units designed similar to this such as the Magneplanars and quite a few ribbon designs. But the laws of physics and mathematics regarding slots is resolute and cannot be broken or altered. The physics of slot mathematics apply equally to sound as well as light and those who would like to pursue this can find a lot of literature in the science libraries of most universities and technical schools.
I am jumping here to the third device, like the Quad, because it exists and has been a very successful approach. All credit should be given to its creator, Peter Walker, as it is a magnificent reproducer. Its only drawbacks are that it is a compromise regarding its radiating field. This lies somewhere between the near field and the far field. If most of you aren’t aware, a cone radiator will have a fall off of radiated energy that is the inverse square of the distance. On the other hand, a slot radiator will have an energy fall off that is just the inverse as there is no square term. The Quads, due to their multiple radiating rings, fall somewhere in between.
Therefore, you cannot get too far away from the Quads and it is absolutely mandatory that they be off the floor. I have found that the most optimum distance is around 6 feet in front of them and boy are they glorious. But like all things, they cannot turn a crummy recording into a peach. I’m not going to say much more about them here because to me, they are the best thing that has come along yet even though there are some drawbacks.
The third shape factor is as mentioned earlier, a truncated scalar form. To describe this is much easier than trying to build it. To digress, most people (and virtually all speaker designers) seem to think that flat frequency response, low distortion and wide bandwidth are the primary goals to achieve. Unfortunately, they are all wet. It just ain’t so. These are SECONDARY importance items. The first order importance items are as follows. Number one: equal speaking time at all frequencies between about 200 Hz and 20kHz. That is, the time response should be optimized so that the leading edge of any frequency waveshape should arrive at the ear at exactly the same time. Easier said than done. Of course, any planar type radiator has a huge advantage in this regard. There is simply no cone loudspeaker system on earth that can equal this—period. However, the second first order importance item is the real killer. It is as follows and all speaker designers should pay particularly close attention to this. The radiation angle should also be as equal as possible between the aforementioned frequencies of 200Hz and 20kHz. Trust me when I say that this is no easy feat to accomplish. As a matter of fact, this is singly the most impossible attribute to achieve. So far, the Quad comes as close as any speaker ever has but it still has a quasi-comb filter effect due to the multiple diameters of the radiating rings. But it is still the best that has ever been done in this regard.
Now, how could this be made better is the question. A six-foot tall scalar triangular radiating membrane, which is slightly curved at the base, would be ideal. In this, only the top 5 feet would be the radiating element. The width of this membrane would be such that the radiating angle would be approximately 1-1/2 radians or about 80 degrees. Many years ago Dick Sequerra produced a speaker called the Metronome. Although this first device was a bit crude and the proportions were far from correct, it did have some absolutely amazing properties that I have yet to EVER experience in another loudspeaker since (save for a later version of his metronome which was much taller). Assuming that one has an optimally recorded event (not an easy task) there are two distinct things that a loudspeaker should do. Firstly, one should be able to walk right down the middle on the centerline of the speakers while closing one’s eyes (therefore letting one’s ears do the work) and as one approaches the speakers one should get the experience of walking right smack up to the orchestra. WOW. Folks, I have experienced this and it is enough to send goosebumps right up your spine. This is the kind of thing I’m talking about when I state that the radiation angle must be constant. A couple of recordings that absolutely knocked me out with this effect were Ken Kreisel’s recordings from M&K Sound. Ken is one very sharp recording engineer whom I might add has probably the finest set of ears in the business. He KNOWS what things are supposed to sound like.
The second effect that is an acid test is as follows. As you are walking down the centerline of the speakers, there should be a point where you are absolutely dead center between the two speakers. At this point the sound should jump right straight up over your head. This should actually be obvious as there is a point right between the speakers where the lateral plane now becomes vertical. This is the second effect that the Metronomes accomplished. If one could imagine this loudspeaker being another 2 feet taller AND being planar instead of using cones, the sound would be truly the most spectacular ever produced. Ah, but there is a fatal fly in the ointment.
Those of you who read my first two installments might recall the situation of my house verses two of my friends houses wherein each of us had our 6 recordings that sounded great in our place but awful in the other two places. So, what’s going on here? Simply put, the microphone placements, coupled with the loudspeakers in the playback environment are not mathematically adding up right. The angles (or vectors) are not aligned correctly for the majority of recordings. I thought long and hard about this and came up with a solution, which is not necessarily practical or cost effective. However, it is doable and someday someone might attempt this. It is as follows: design a mechanism that I call “SpeakerTrak”. What this would be is a platform that would allow the speaker to be moved left or right, up or down, rotated, tilted, and moved forward or backward. Obviously this contraption would have to be extremely solid and produce NO mechanical vibrations of its own. Most importantly, it would mandatory that this be remote controlled. Therefore, one could adjust the speaker from the listening position. Incidentally, the electronics could also have a microprocessor with memory so that once a position is located for a particular recording, the position and settings could be stored in memory so that every time you played that recording, the speakers would automatically align themselves to the correct position.
A device like this might cause some consternation in a lot of living rooms however, this is the only scientific avenue that could achieve the correct acoustic results. This so-called platform could also be located on the wall or hung from the ceiling. Imagination prevails here. Do you think that I’m nuts for proposing this kind of a solution? To be honest, I have over one thousand recordings and I think about only 30 or 40 of them are really listenable with my present setup. If only I could MOVE the speakers for every recording.
Now let’s tackle another acoustic problem which involves home theater. As I have mentioned previously, having a TRUE center channel situation is the most desirable thing to achieve. BUT doing it correctly is easier said than done. Firstly, a processor with the correct algebraic solution is mandatory. This will be available shortly. When incorporating this scheme, it is absolutely mandatory that the (necessary for stereo) cross-coupled error signals be eliminated from the opposite channel. You can get a feel for what I’m talking about when listening to headphones, as there is obviously NO cross-coupled error signal. The biggest problem is that the overwhelming numbers of center channel speakers that are available are just positively awful. They are designed to physically “fit” in with the screen (mostly large TV) and as such could not (should not) be considered as hifi reproducers. Enough said about them. What one must do in order to have a decent “musical” situation is to get the screen as high off the floor as possible so that a “decent” loudspeaker can be located under the screen and still be tall enough to provide accurate reproduction. You must bear in mind that with this kind of setup the center channel now becomes the MOST important speaker and should be of the highest possible quality.
Also, FORGET ABOUT THE REAR for music reproduction. It is a red herring and I predict that this potential multi-channel recording baloney is going to end up just like Quad sound of the `70s, that is, a debacle. This situation is truly a fool’s errand wherein, just like in the `70s, marketing idiots seem to be spearheading this nonsense. I was right about the nonsense back then and trust me, I’m right about it now. The rear speakers should only be used for the surround in home theater where the effects can be truly spectacular. But in music,THERE IS NO REAR. So there you have it. The very best playback system would consist of the following: left, right, and center speakers with high and low center speakers with a subwoofer. And, there should be provision to adjust the physical properties of at least the three main front speakers. And finally, there should be the ability to “dial in” your preferred listening position. A tall order, huh? Well, we’ve had 45 years of chickens all scattering in a myriad of directions and very few roosters. Somebody now needs to step up and scream COCKADOODLEDOO.