Suppose that Leonardo da Vinci had known about the hologram. And suppose he
was aware of a technical clue that he could put onto his paintings allowing
future generations to see them as three-dimensional images in untarnished colors,
once they developed the technology for proper reproduction. And suppose that
because the technology of his time was not sufficiently advanced, Leonardo
realized he would never be able to see these images as future generations could.
Art style or artistic intention aside, what would be your reaction if he had known about the possibility of these things and chose not to bother adding this clue, even though it was otherwise invisible on his paintings?
Before you chastise the memory of da Vinci in this hypothetical circumstance, consider this: Right now, today, we can record subsidiary information on our master recordings which will allow future generations to reproduce that sound with an accuracy denied us ... and we are not presently doing it.
Rosetta Stone
It is conceptually quite easy to understand what we could do to increase the future value of our present sound recordings. The information we could leave has been likened to that of the Rosetta Stone (1, 2). Found in 1799 near the town of Rosetta in what is now the Arab Republic, this stone bore parallel inscriptions in Greek, demotic, and Egyptian hieroglyphic characters. Since Greek and demotic were known to scholars, this made it possible to decipher previously unfathomable Egyptian hieroglyphics. If our audio Rosetta Stone contains adequately recorded reference sounds which will be known to future technologies, as well as the sound of the performance, then, like that ancient cipher, this reference can be used to remove the veil of years and technology.
Evolving Technology
Within a decade, the recording and processing of sound will swing almost exclusively over to the digital realm.
Several decades after that, the methods by which recordings are made, processed, and reproduced may differ considerably from anything we can now envision. Audio, for example, may not survive as a separate art. Simply "listening" to something may not be considered adequate, no matter how accurate the sound may be.
Regardless of such considerations, it is apparent that the methods of recording and reproduction which we now use will eventually be replaced by something else. I have a personal feeling what one of those "something elses" may be. But nobody can really say what will happen. We see the future only through the window of our imagination . . . and that window is often distorted by our knowledge of present capabilities.
But we do have greater assurance that whatever future generations may do, they probably will not change the laws of nature. This means that if we can preserve sound in a way somehow linked to the laws of nature, we can leave a legacy of great value.
Let me point out that, like my hypothetical da Vinci, we do know how to intercept a sound field in such a way as to contain the major ingredients of an acoustic hologram of that sound field. We can record it, but the rub is that we do not yet have the technology to reproduce it. And the interesting fact is that we can do this for many of our present recording situations without effecting the present product in any way.
In addition, we know what actions we might take to provide test signals on our recordings that can be used to correct many of the inherent technical limitations in our present archive preserving techniques.
Think, then, what impact it would create if many years from now the technology was developed to recreate the sound field corresponding to the intercepted dynamic diffraction pattern of some of our present performances. Imagine the legacy we could leave if we allowed some future audience to hear the sound of a legendary artist performing in a hall that was no longer in existence.
Past and Present
Perhaps the sound reproduction technology of that future era will bear no resemblance to anything we now consider "necessary." They may chuckle at the history book descriptions of our two-channel "stereo" systems and wonder how in the world we could ever tolerate such distortion. And there might not be many of those ancient magnetic or disc reproducers available in the museums which could be used to play the archive records.
"Imagine Hearing a Legendary Artist Performing in a Hall That was no Longer in Existence."
But if they had dutifully transcribed the record-to-computer bank-to "whatever," keeping all Rosetta Stone information, much as we do today to preserve 78s by putting them on magnetic tape, they would have it available when the time came. And they could put it into whatever format of reproduction was then in vogue.
And, if they were smart, they would realize that their "present" would again become somebody else's "past." And they would not throw away those old "records" once the wiggles were traced into computer "whatsits." But they would make every effort to preserve our scratches and magnetic domain orientations so that "future future" technology could squeeze even more out of them.
If that sounds strange to you consider this: In thinking about the concept of an audio Rosetta Stone and what sort of signal previous generations might have inadvertently left on old acoustic records, M. Gerzon, with the mathematical institute of the University of Oxford, postulated that there may be a vestigial stereo signal present in them. (3) The vocalist generally stood directly in front of the recording trumpet, while the stronger instruments were placed off to the side. Although the intent was to produce either a vertical or lateral groove modulation, there was a tiny off-axis motion of the stylus due to acoustic waves coming from the side. Such a component was actually found when an early acoustic recording was traced by a stereo cartridge. It is not clear whether this inadvertent stereo signal can be reasonably lifted off the original recordings by present computer technology, but had recordists of several decades ago "preserved" the signals by simply transcribing the acoustic record monaurally to electrical transcriptions or magnetic tape, then thrown the old records away, this phenomenon would never have been preserved.
Two Aspects
There are two aspects to this concept of leaving a Rosetta Stone. One aspect deals with leaving adequate information so that future processing can remove, or reduce, not only the existing technical problems in our recordings, but correct for the inevitable deterioration effects while in archive storage. The other aspect deals with preserving the spatial configuration of the original sound field through providing enough information to generate the equivalent of a dynamic diffraction pattern.
In some ways both of these aspects can be addressed by the same type of subsidiary signal which we can leave.
A key to this is to use some type of acoustic-oriented signal which depends upon first physical principles and can be duplicated by future generations.
Let me give an oversimplified example: If, prior to a recording, we walk through the hall and clap our hands once every several seconds, we are exciting that hall with a localized impulsive source which migrates through the recording environment. A recording of that sound will contain the room response as intercepted by the microphones. If careful annotation is made of the room geometry and location of each clap, we have an acoustic spatial record of the room. If a musical performance is made in that same room and with those same microphones, we can theoretically combine them to determine the spatial position of each instrument in the performance. Even without the recorded hand claps, we have a synthetic aperture situation when there are a number of microphones, spatially separated, which can pick up the sound of the same instrument. The hand claps give us needed polar patterns of the microphones and the nature of reverberation species within the recording environment.
Of course, it is not perfect ... and it is also a pain in the gluteus maximus.
But if this Rosetta Stone signal goes through all of the equalization and processing that the performance does, and if it is subjected to the same storage conditions, and if future generations know how to duplicate a hand clap, then they can crunch on all this data to reconstruct an actual sound field far closer to that of the original recording environment than anything we can get from this same recording today.
I only use the hand clap as an illustration of a type of signal. Actually, a hand clap is far from ideal. But a hand clap is better than nothing at all which is what we now have.
There are many technical factors that must be addressed if we want to do a good job of inscribing an acoustic Rosetta Stone. The fundamental consideration is to leave some type of signal which can be used to "measure" the technical limitations of the recording medium. Phase-coherent frequency sweeps at different intensity levels would be nice to have.
The next thing would be a signal that could be used to determine the microphone and pickup limitations, once the recording medium aberrations were removed. Even a dumb thing like a child's spring-steel clicker toy could be used to get started. Go up to each microphone and snap the spring for near field and high sound pressure level characteristics. Then back off to the location of each musical instrument that is to be recorded and snap again, always being careful to note where the sound is relative to the microphone. Since it is a test recording we can speak out loud and say where it is located as we walk around.
Photograph the hall and mike placement. Make a B & K microphone recording of the sound of that clicker so future generations can get close to the original sound even if they did not know how to physically duplicate the clicker. And be sure to put it all together as an archive package.
I have no intention of going into greater technical detail on what type of signal we could use, nor do I wish to haggle about the fact that many of our recordings are not intended to represent a cohesive spatial-temporal performance. What I want to do is stimulate thought about the concept to free our mind from the prison of the present.
The Hall in a Can
The acoustic wave propagating outward from each source of sound can be thought of as a signal illuminating the listening environment. The reaction of that environment to this acoustic illumination gives rise to the sound field. If we know what the source of sound is and we intercept the resulting sound field, we can use that to recover the "sound" of the environment.
Think what that means. We can conceptually preserve the acoustic "essence" of a great hall (4, 5). And it does not take as much effort to do this as we might imagine. Since an illumination function has a predictable space-time relationship, we can determine most of it by intelligently intercepting selected parts of it. We really should record both the sound pressure (a scalar) and particle velocity (a vector) at a number of different locations.
But if we only had one part, such as pressure, we could crunch to determine the probable velocity distribution.
Once we had the "hall in a can" there is no necessity for performers to go to that hall in order to produce a recording "there." Regardless of our esthetic or professional opinion on the ethics involved, it is going to happen some day. All we need is the appropriate acoustic Rosetta Stone. That sure will stretch the old copyright laws to the limit. Who owns the sound of a great hall?
The Intellectual Trap
Of course, when I mention such things I am falling into the common habit of restricting the future to those things I know can be done today. But future technology will take directions that we cannot predict. For that reason if we ever do get around to recording the subsidiary information which can be used as a Rosetta, we must try to document it as thoroughly as possible.
I close this little mind-stretch discussion with a personal observation .. . there is a fascinating intellectual trap that gets sprung once we know it is possible to leave a Rosetta Stone for future technology. Once we know that something of value can be done for the future, we can never remain intellectually content in the knowledge that it is not being done at present.
References:
1. R.C. Heyser, "A Rosetta Stone for Audio?" (editorial), /. Aud. Eng. Soc., Vol 22, No. 4, p. 251 (1974).
2. D. Davis, "A Rosetta Tonet" (Forum), J. Aud. Eng. Soc., Vol 22, No. 6, p. 430 (1974).
3. M. A. Gerzon, "On a 'Rosetta Stone' for Acoustic Recording," (Forum), J. Aud. Eng. Soc., Vol 22, No. 8, p. 638 (1974).
4. M.A. Gerzon, "Recording Concert Hall Acoustics for Posterity," (Forum), /. Aud. Eng. Soc., Vol 23, No. 7, p. 569 (1975).
5. R. C. Heyser, More on "Recording Concert Hall Acoustics for Posterity," (Forum), /. Aud. Eng. Soc., Vol 23, No. 7, p. 571 (1975).
(Source: Audio magazine, Jan 1979; Richard C. Heyser)
= = = =