THE QUEST FOR AUDIO NIRVANA (2/3)
WHAT PRICE GLORY — PART TWO
In the first installment, I discussed some aspects of electronics relating to matters audio. In this second part I will divulge a few unknown or a least obscure things relating to the recording process. Firstly, I would like to reiterate a previous statement: Stereo DOES NOT exist in the real natural world. I really want to drive this point home. All sounds in nature are created in MONO. However, as everyone knows, mono has no depth, no dimension, and no presence. Therefore, stereo becomes a necessary evil that at least provides us with some semblance of the foregoing. It has been up to our ear-brain link to fill in all the gaps of missing or out of kilter information. Even though stereo itself is a lousy format that we have been living with for some 45 years, it didn’t have to be that way. There has been a mathematical derivation that could have been provided for us right from the beginning had the engineering community been smart about the whole process. Unfortunately, it wasn’t to be but, that won’t be for long. Anyway, let me define for you what I mean by the words dimension and presence.
Dimension is a quality that has 3 derivatives. First is the main aspect of stereo which encompasses the lateral or left to right dimension. The second aspect is the front to back dimension. The third derivative is the one that is missing which is the vertical dimension. For the most part we live on a flat earth wherein alll of our lives and the events in our lives occur in the lateral plane. Also, our ears obviously are on the sides of our head and not on the top of our head and on our chin. Our human perception is much more acute in the lateral plane than in the vertical plane. For example, try locating an aircraft WITHOUT moving your head. You will find it to be extremely difficult. Virtually everyone looks up into the sky and tries to locate the plane with their eyes and then, once located, it’s much easier to slightly capture the sound direction.
In stereo recording, all of the information is recorded laterally and there are no extra channels to capture the vertical information. There is some savings grace due to psychological aspect of associating high frequencies above low frequencies. That is why virtually all loudspeakers have the tweeters at the top and the woofers at the bottom. If you want to have a good laugh, try turning your speakers upside down to hear what they sound like. You will be very surprised to hear how weird they sound.
The second thing I want to explain is presence (my definition). This is the aspect which determines your seating position. When I go to an event, I want to sit right down in front—first row. I have several friends however, that can’t stand to be that close. They generally sit about 30 or 40 rows back. Now, if you are a recording engineer, where do you put the microphones in order to satisfy both parties? The answer is: YOU CAN’T. This is a problem of time and not amplitude. You cannot fix this problem with the volume control.
Another part of this dimension problem involves the spacing between the microphones. In the old days of wide spaced mikes (called stereophony) the mids and highs were quite acceptable generally however, the low freqs were really a problem. This occurred because with wide spaced mikes the wavelength difference at low freqs caused multiple eigentones and cancellations in the playback environment. Most recordings tended to sound cavernous and boomy at the low end. In order to solve this problem, coincident mike techniques were developed and employed with the capsules virtually on top of each other with angle spacing of between 110 degrees and 135 degrees. This virtually solved the low frequency problem but now introduced problems at the high end. For example, if the capsules are in effect say about 1 inch apart, then that distance represents a wavelength of about 10kHz. When played back with the speakers say 8-10 feet apart, a gigantic hole is produced wherein one must find the exact “sweet spot” in order to listen.
There have been other techniques that have been used to help with these problems such as the MS and crossed figure 8’s (a Bert Whyte specialty). But the problem really involves time and distance and not amplitude.
A few decades ago when I was faced with these kinds of problems trying to record my Steinway grand, I was indeed frustrated. After pondering this for quite a while, I devised a totally different concept in mike technique. Firstly, I used a MONO center channel that went through a low pass filter with a cutoff at 250 Hz. I then placed a pair of mikes (in stereo) on either side that covered the frequencies of 250 to 500 Hz. These of course fed two bandpass filters which went to a summer on each channel. The distance on either side of the center mike was exactly 3 feet. I placed then, a 2nd pair of mikes further outside to cover the frequencies of 500 to 1000 Hz. These were placed exactly 1.5 feet outside of the first pair and their outputs went through a corresponding bandpass filter and into the summing circuit. I then placed a 3rd pair of mikes exactly ¾ of a foot to the outside of the 2nd pair which covered the frequencies of 1000 to 2000 Hz with the respective output going through bandpass filters to the summing circuit. Finally, I placed a 4th pair of mikes exactly 4.5 inches outside of the 3rdpair which covered the frequencies of 2000 Hz and up wherein their outputs went through high pass filters to the summing network. I would liked to have used another pair in order to make the last band 4000 Hz and up but I didn’t have any more mikes available.
To say the least, the results were absolutely astonishing. Where I had “mush” before where everything was very hard to distinguish, now everything was crystal clear and the piano had absolute focus and position. Obviously, this is a very difficult procedure however, I wanted to prove a point. Correct mike location and technique is THE absolute most critical aspect of recording.
I’m sure that you can comprehend that a 2 channel system can only approximate the exact position of a source of sound in space. If you were to compare this with radio direction finding which uses TRIAGULATION of three receivers to locate a source, it’s easy to understand that a 2 channel attempt is going to fall short of the mark. Another thing that should be understood is that on an absolute basis, there is actually very little LEFT ONLY and RIGHT ONLY information. Virtually all of the primary acoustic information falls basically in between the loudspeakers. Yes, I can hear the cries of “not true” coming from all those who believe that the sound stage extends beyond the speakers to the outside. This is of course true but, it is due to the reverberant field and not the stereo composite signal itself. Notice also, that I said very little and not none. As a matter of fact, the most important information, for the most part, is located right in the center or mostly in the central field. Obviously there is no speaker there. This is what I meant by our ear-brain link having to fill in the gap of missing information.
A long time ago Nakamichi tried to improve recordings by adding a center fill microphone jack on their cassette decks. Sometimes this helped but most of the time it didn’t mainly because the wrong microphones were being used. In order for something like this to have even half a chance, a thorough understanding of the “vectoring” of microphones is necessary. With this 3 mike scheme simple cardioids just will not work because there isn’t nearly enough isolation between the mikes. What is needed here is super or even hyper cardioids and they must be set up very carefully in order to make sure that there is as little leakage between them as possible.
Now lets get to the meat of some suggestions and proposals. For the last 45 years or so, we have not had much choice in recording techniques because we were limited to only two channels. Of course I assume that everyone remembers the absolute disaster that befell us in the 70’s. I’m speaking of course about “quad” sound. Anyone with 5 cents worth of brains could have or SHOULD have seen the lunacy of this approach. As it turns out, I and others who tried to scream loudly, were ignored. But unfortunately, after the millions of dollars were wasted on this fool’s errand, we engineers had the last laugh. Sorry ‘bout that. We are now faced with another idiotic adventure by those who are proposing multichannel recording???? We haven’t conquered (and never will) 2 channel recording and some people want to use 5 channels? This is absolute nonsense for the following reason. All primary musical information COMES FROM THE FRONT. It doesn’t matter what event you go to whether a concert or a club etc., all of the musical information is in front of you and not behind you. So the question becomes: is there a better way? The answer is, you bet.
We have staring us in the face an incredible opportunity to have our cake and eat it too with the introduction of DVD audio discs. These discs have about 8 times more capacity than a regular CD. Why not make proper use of this gift horse in the mouth. For example, the 5 channels could be used in the following manner. Three for the left, center, and right and a fourth for a high center, and a fifth for a low center. This would give us that missing vertical dimension. The playback speakers for the high and low center would be placed accordingly and only have to be reasonably acceptable in quality. To go a step further, many sets (say four sets) of mikes could be located at intervals of say about every ten feet with the first set being close miked, etc. Sound wild? Well wouldn’t it be nice if you had an adjustment control on your preamp where you could actually dial in your preferred listening position. Incredible concept. Trust me, it’s doable. All it takes is some industry leaders with GUTS.
Finally, a little anecdotal story is in order to prove some of the things I’ve been talking about. Many years ago I did a round robin listening experiment with a couple of friends. I had them over to my place wherein I had selected 6 recordings that I considered absolutely spectacular relative to miking, tonal color, etc. We played these recordings and indeed they sounded tremendous on my system. We then went over to the first friend’s house and my six records sounded absolutely LOUSY. He then pulled out six of his favorite records which indeed sounded glorious on his system. We then went to the third friend’s house and lo and behold, both my six and my first friend’s six recordings sounded just terrible at our third friend’s house. But then he pulled out HIS six favorite records which sounded incredible on his system. We then came back to my place and wouldn’t you know it, both of my two friend’s six recordings (each) sounded like absolute crap on my system whereas my original 6 sounded great. The point of all this is to show all of you just how much we are still in the stone age and just how far we have yet to go. It’s going to be a long and bumpy ride but hopefully someday this industry can clean out the cobwebs of egotism and get down to some truthful scientific discipline and come up with credible answers to this very enigmatic audio puzzle.