Down with Dynagroove! Page 4
The April 1964 issue of the Journal of the Audio Engineering Society carried a detailed article by RCA's Dr. Harry F. Olson (footnote 3), which shed considerable new light on the controversial RCA Victor Dynagroove system. On the basis of this, plus some other recent articles on the subject, The Stereophile is now obliged to retrench on one point with which we originally took issue, but we are even more firmly convinced that most of the other aspects of the system represent a giant backward step in the recording art.
The predistortion phase of Dynagroove—the adding of complementary distortion to offset the effects of stylus tracing distortion—appears now to be based on sound engineering principles, even though the amount of predistortion that is added is predicted on the normal amount of tracing distortion produced by a 0.7-mil stylus, a size that has been generally abandoned by the high-fidelity industry. The IM that results when the predistorted groove is traced by a smaller stylus is, however, still less than that obtained when the same stylus traces a normal groove. This has been proven mathematically and demonstrated subjectively.
We applaud RCA's extraordinary efforts to eliminate acoustical problems in their recording studios (even though "well engineered" studios have a reputation for being characterless and antiseptic-sounding), and we are impressed by the measures RCA took to avoid overmodulation of their discs. But we continue to be appalled at the reasoning behind the "Dynamic Spectrum Equalizer."
The three response curves that were included in Dr. Olson's article, represent the tonal compensations that are added automatically to Dynagroove discs (and 4-track tapes) for three differing intensity levels of orchestral playing, ranging from almost full-orchestra volume (100dB) to quiet solo-instrument volume (40dB). These curves are added to the RIAA recording curve, so when a Dynagroove disc is reproduced via the "industry-standard" playback curve, the overall response of the signal will be as shown, according to the volume at which the orchestra is playing. At no time is the resulting playback response even remotely linear.
According to Dr. Olson, these tonal corrections are added to offset 1) the low-frequency masking effects of background noise in the average listening room, 2) the fact that home listeners listen at lower-than-concert-hall volume levels, 3) the tendency for an average listening room to exaggerate bass response, and 4) the tendency for the average loudspeaker to radiate its treble in a narrow beam rather than a broad field. Dr. Olson then goes on to explain that the loudspeaker used as RCA's "reference system" in a "typical" room was an RCA LC-1A, whose performance characteristics, as reported in the article, reveal neither a low-end rise nor a tendency to beam treble.
All of which leads us to the conclusion that people who like the sound of Dynagroove recordings must either have very, very average systems and listening environments, as described by Dr. Olson, or are very poor judges of natural sound.
Incidentally, one of the basic requirements listed by Dr. Olson for the Dynagroove system was that the recording microphones should have response characteristics falling within...a response deviation of almost 5dB—a total variation of nearly 10dB—with all of the deviations as bass attenuation and treble accentuation. We have often remarked, half jokingly, that the engineering mind deems it better to be up 10dB at 10,000cps [hertz] than down a half a dB. Suddenly, we don't think that remark is very funny.
—J. Gordon Holt
Footnote 3: Then Director, RCA Acoustical and Electromechanical Labs, Princeton, NJ.