DIGITAL RECORDINGS, A TRAGEDY UNRIVALED IN
THE HISTORY OF ART
©1985 Mark B. Anstendig Foreword
Since
this paper was completed, many who truly understand the permanent flaws of the
current, industry-adopted digital technology have been won over by the argument
that, although this digital technology is flawed and can never achieve the
accurate results of well-reproduced analog recordings, it still constitutes a
remarkable improvement over most people's record-playing equipment. That is an
insidious and dangerous argument. It loses sight of the crucial point that,
while analog recording is able to capture all of the important sound
information, the digital system that has been adopted by the industry can never
do so. And the information that CD does not reproduce is the most essential
information in the reproduction of the original musical experience: the
fine nuances in the flow of the sounds that convey the emotional-expressive
experience.
With
analog recordings, the essential information, including all of the dynamic
fluctuations, is preserved on the record. It is the playback equipment that is
usually incapable of retrieving it. The record-owner can always upgrade his
equipment and learn a few basics of correct sound-reproduction in order to get
back to the real experience of the original performance. But, with CD discs it
is impossible to reproduce the original musical experience. The owner of CD
discs can never retrieve that last, most important missing measure of the music
no matter what he does, because it simply is not there in the original
recording. What is heard is always interpretively different from the
original performance.
The
built-in failure of current digital technology is that the sampling rate is too
low. There is no technological argument against that point. Every
manufacturer knows about and understands the limitations of the adopted
sampling rate. A faulty technology is, therefore, being knowingly rationalized
and foisted on the unwitting public. Various techniques (filtering, multiple
sampling, etc.) are used to smooth out the holes in the flow of the sound due
to the insufficient sampling rate. But these techniques merely fill in the
spaces between samplings. None of them replace the missing fluctuations in the
sound that were lost in the recording process. The enclosed diagram makes that
clear.
Even
the wonderful "specs" supplied by the manufacturers are illusory. In
careful analysis and comparison with analog, few of them hold up. But, more importantly,
all published specs are measurements of isolated, static moments in the flow of
the sound. None measure the accuracy of the reproduction of the dynamic flow of
sounds in time (indicatively, digital hopelessly fails the measurement of
so-called "square waves”, the most widely used measurement that attempts
to analyze the flow of sounds in time).
Manufacturers
of sound-reproducing equipment have just as much obligation to preserve the
values of the arts they are recording as the musicians themselves. That digital
is an improvement over faulty playback of analog sound-reproduction in no way
justifies its manufacture. In music, the result of the inadequate sampling rate
of digital is that the expression of the music is permanently degraded
and changed into something different from, and inferior to, that of the
original performance. The crowning glory of any of the sonic arts--the
expressive content--is, therefore, destroyed. There is no way whatsoever that
such a result can ever be justified.
When
well-reproduced digital recordings are compared to well-reproduced analog
recordings, there are clear differences to be heard. That many of the finest
technological and musical minds have not heard these differences and have
already accepted and even praised digital sound is a failure of human
perception that constitutes a human tragedy of a magnitude equal to the
artistic tragedy described in the paper that follows.
The
introduction of digital recording processes before they had been perfected is
destroying the interpretive content of old and new recorded master
performances. This obliteration of thousands of great artistic performances,
many of which are the aural equivalent of the Sistine Chapel, is a tragedy
unparalleled in the history of art. But, since the public is not familiar with
the interpretations or the actual sounds of the original performances, it has
no way of knowing how much of the interpretive nuance is missing and therefore
has no way of recognizing and evaluating the enormity of this tragedy.1
For
a century, the public has been listening to distorted recordings, broadcasts,
and sound-reinforcement that reproduce neither the tone quality nor the nuances
of the original sounds. As one system was improved to the point where the
achievement of some accuracy in the final playback was nearly in sight, a new
system was introduced that necessitated beginning again almost from scratch.
The only logical change was the replacement of 78 rpm records with the
long-playing record. When LP recording techniques were already quite excellent
but playback techniques still needed to be perfected,2 stereo was
introduced. Stereo is, unfortunately, merely an added effect and by no means a
necessity for reproducing music. But it did create a situation where the manufacturers
could sell more expensive components with nearly double the number of parts.
Since the recording and retrieval of two separate signals on one record created
extreme technical demands, especially on the record-playing components
(turntable, tone-arm and pickup), a whole new technology had to be developed.
At
least old-fashioned analog stereo recordings are capable of capturing all the
important sonic information in a musical performance. Stereo records achieved a
technical level where essentially all that remained was to perfect the playback
equipment and certain techniques of compensating for changes in the sound that
are inherent in the reproduction process. But, in 1979, when stereo playback
systems were still only approaching the ability to retrieve accurately all of
the information on the records, the industry introduced a form of digital
technology that will never be able to record a musical performance accurately,
not to mention play it back.
The public should avoid:
1) all CD discs,
2) all digitally recorded phonograph
records,
3) all digitally re-mastered analog
recordings.
Incapable
of preserving all of the sonic information of the original musical performance,
these digital recordings will never be able to reproduce the important subtle expressive
nuances that are the mark of great musical performances.
Digital
recording processes, which could eventually have become the ideal recording
media, were introduced long before many of the necessary techniques were
perfected. The main problems are too low a sampling rate and too undependable a
storage medium. Tragically, against the will of many knowledgeable experts,
this imperfect digital technology has been universally adopted as the norm for
the entire recording industry.3 With analog records, it was possible
for small, conscientious companies to bring out technically better records made
with different systems, such as "direct-to-disc.” But, since the major
recording companies all use the one digital technique and better systems would
demand completely different playback equipment, it is impossible for other,
conscientious companies ever to bring out a satisfactory digital technique
unless the whole industry adopts it.
The
public has not recognized the failings of digital recordings because they have
never heard how good their analog records are. But, while all the information
is preserved on old-fashioned stereo and mono records and techniques have
finally been developed to retrieve that information fairly accurately, the
information that is missing on digital recordings can never be retrieved, no
matter how much the playback equipment is perfected. Comparisons of analog and
digital recordings (on finest playback equipment with the sound properly
equalized4) clearly reveal the superiority of analog recordings.
Among other things, the expression is different because digital cannot preserve
the dynamic fluctuations in which the expression resides: digital is deader,
lacking lifelike qualities; the high frequencies are harsher; low-level sonic
information (ambience, overtone structures, etc.) is missing; and the sound
loses most of its vibrancy.5
The
reason the faults of digital are not immediately apparent is due to the faults
of most record-playing equipment. The public is no longer able to differentiate
between impeccably played and merely run-of-the-mill musical performances
because music-making in general has deteriorated for decades due to the
influence of poor record-playback equipment which, like digital, was incapable
of reproducing the interpretative characteristics of the great recorded
performances.6 Until the advent of digital, the recordings, even old
78's, did preserve the fine expressive information and today, in 1985, those
persons with today's state-of-the-art record-playing equipment and equalizers
can now hear it. But, with current digital recordings, that information is
irretrievably lost because currently used digital techniques are incapable of
preserving the nuances during the recording process itself. Ironically,
digital, if perfected, would be the ideal recording medium, but the agreement
on an industry norm effectively keeps a better system from being brought out.7
The
adoption of these faulty digital recordings is being accelerated by two
industrial developments: old, excellent analog records, are being remastered on
digital equipment and many radio stations are switching to the exclusive use of
CD discs because of their ease of operation. The public has already been
conditioned by bad record-playing equipment to accept a distorted perversion of
expressive qualities. Broadcasting digital exclusively is a monumental mistake.
It may well mean the end of fine music-making as it is bound to cement in the
listening public and musicians alike a wrong idea of musical expression and the
expression of emotions is the most essential and human of musical values.
The
re-mastering of great, old analog records is a wholesale destruction of fine
art unthinkable in any other artistic field. Many recordings by Otto Klemperer
and the Philharmonia Orchestra, for example, are as great an artistic
achievement as the Sistine Chapel of Michelangelo. They are the result of long
lifetimes of study, preparation, and experience by conductor and orchestra
members alike. The problems that were overcome in these performances equal the
problems Michelangelo had to overcome in painting the Sistine Chapel, and the
richness, subtlety, and precision of execution are second to nothing else on
this earth. For these masterworks to be re-released to the public in re-recordings
that are re-mastered by digital processors, as is currently being done, is the
same as destroying all of the extraordinary expressive details in the Sistine
Chapel paintings. It is an artistic tragedy of unthinkable proportions.
The
Anstendig Institute's papers on digital and those on the problems of
sound-reproduction in general are available free of charge from the institute.
1 The Anstendig Institute's other papers on digital explain in
detail why it is a faulty recording system.
2 Stereo is only one of many possibilities of creating
directionality and the impression of spatial dimension, none of which can
recreate the original space. Our paper, "Stereo; A
Misunderstanding" explains the fallacies of stereo sound reproduction.
3 That the shortcomings of digital processors were known to the
manufacturers when they introduced the system is clear in that all digital
processors contain, in addition to complicated noise filters, special circuitry
designed to guess at and attempt to replace what has been missed by the
too-slow sampling rate or by dropouts in the storage media. None of this would
be necessary with a high enough sampling rate and an adequately dependable
storage medium.
4 Our paper "Sound Equalization in Relation
to the Way We Hear" explains the need to equalize all reproduced
sound.
5 The failings of digital sound have been well documented in the
serious audio publications, such as "The Absolute Sound", that are
independent in that they do not depend on industry funds for advertising. One
example of a telling fault is that, above 7000 Hz, square waves (bursts of
sound) are reproduced as steady tones; in other words, a 100% error rate. And
the cause of this problem affects the music at lower frequencies, just less
noticeably.
6 The points made in this paragraph are elaborated in The
Anstendig Institute's papers on sound-reproduction and acoustics.
7 To be perfected, digital must, first of all, have a sampling
rate of 300,000 cps or more (the current sampling rate is only 44,000).
The Anstendig Institute is a non-profit, tax-exempt, research institute that was founded to investigate stress-producing vibrational influences in our lives and to pursue research in the fields of sight and sound; to provide material designed to help the public become aware of and understand stressful vibrational influences; to instruct the public in how to improve the quality of those influences in their lives; and to provide the research and explanations that are necessary for an understanding of how we see and hear.
|