©2008 Mark B. Anstendig
While the authors of the attached article complain about the
deterioration in sound over the last decades, they still miss the fact
that the expressive/detail of the sound is lacking, not just the
dynamic range and whatever else they mention. Early digital recording
systems could not capture the full nuances of sound as it flows in
time. What suffered most was the expression contained in the original
musical performance. That was lost in the recording, as described in
previous Anstendig Institute papers.
The state of the art in recording still remains the Sao Win latest
analog record-playing pickup cartridge.
The following paper merely comments on and explains our view of the
Rolling Stone article, linked below.
Three decades too late a Rolling Stone article acknowledges the deterioration/disaster that we predicted in sound reproduction but for the wrong reasons.
The Anstendig Institute has predicted the demise of decent recorded sound for over three decades. The accompanying article deals mainly
with one important part of the problem: dynamic compression.
Originally, before digital recording, the only real problem of analog
records and tapes was a limited dynamic range. Analog means were not
able to produce the full dynamic range of most live performances.
Therefore, some judicious "Gain riding" was used to fit the dynamic
extremes of volume onto the tapes or discs. This was usually confined to lowering the volume on climactic passages, but also sometimes used to raise the volume of very soft passages as well.
The Anstendig Institute has always advocated and used reverse gain
riding during playback to restore some of the lost volume extremes:
while listening, one of us, usually Mr. Anstendig, uses the volume
controls manually to expand some musical climaxes and then reduce the
volume when the reduced-volume passages are over. Soft passages that
have been made louder are less bothersome, but we also try to
compensate for these distortions when they are noticeable.
It was long ago proved by Fletcher and Munson in the famous Bell Laboratories tests, that when the overall volume level is changed, the equalization (timbre) of the way we hear the sound changes radically. The famous "Fletcher-Munson Curves" graphically plot the differences in the way we hear the same passages at different volume levels. Changing the overall/whole volume level of a recorded performance (or any of the inner voices/voicings in the sound) changes the timbre of those sound, in effect ruining /wrecking the way the music or anything else sounds. The sound is muddied in the mid-range and low frequencies and/or excessively brightened in the irritating high frequencies. Instruments and voices sound different and unnatural. The sound becomes irritatingly grating when the high frequencies are emphasized and muddy and unclear to the ear when the mid or low frequencies are emphasized. Usually there results a mess of all those results. Even in mainly talking-only recordings, like films and TV shows, diction is messed up and becomes hard to understand. The whole artistic-emotional experience of the program is compromised. The only cure for gain riding, even for the smaller, more judicious amounts of compression originally used in analog recordings, is to use an equalizer (a more capable tone control) to eliminate (equalize) the timbre distortions from the volume changes and restore something close to the timbre of the original performance or live event.
The messed-up volume levels described in the accompanying article takes this problem to ruinous heights, where images are relied upon for the emotional and artistic experience of the viewer, and the sound played a secondary role. That is horrendous, because sound is the more powerful of the two higher senses, and the effects of sound/hearing override those of sight. Therefore, the main artistic/ emotional content is wrecked and ruined to the point of no longer being even close to
that of the original performance, be it purely musical or theatrical
mixed sounds and images.
The worse offenders in this problem are not even mentioned: they are the commercials in TV and radio. In profound ignorance, it has been common practice even before digital sound to substantially raise the volume of all commercials, ostensibly to attract the viewer's/listener's attention, but without re-equalizing the sound to make it match that of the lower-volume program content. The result is really nasty distortions of the sound in all parameters, which make the sound in those commercials irritatingly grating and truly annoying. Even loudspeaker manufacture has suffered from this problem, because speaker manufacturers often make their speakers deficient in high frequency information, in order to try to combat the irritating nature of these volume-expanded passages. And dull-sounding speakers with deficient dynamic range, freedom and accuracy, abound and are usually preferred, while accurate speakers with full dynamic capabilities are viewed as annoying. The whole process actually works against the effectiveness of the commercials, as they end up irritating the viewer/listener as much as attracting their attention.
And all this can probably never be corrected or lessened, because there is too much sound out there with these problems, and none of it has the same amount of the problem. It could never be sifted through and corrected.