Down with Dynagroove! Page 2
RCA Victor terms their Dynagroove process "the most significant advance in the recording art since the introduction of the LP." Significant it may well be, but whether or not it is an advance is another question altogether.
In response to a telephoned query, RCA Victor's recording department filled us in on some details about their new Dynagroove recording process. Allowing for the possibility of some confusion on our part, here is the story we get.
In a nutshell, Dynagroove is a process for correcting automatically certain "inherent" deficiencies in the tracing and the aural balance of disc recordings.
A cutting stylus, having chisel edges, can cut modulations that are smaller than the radius of the playback stylus. In playback, the stylus cannot trace these modulations, but simply bounces over the high spots, causing distortion and rapid disc wear. Dynagroove's computer modifies the signals that would normally create these untraceable modulations, and converts them to modulations that the playback stylus can trace, and which elicit from the pickup signals similar to those it should produce from these signals. This sounds like a good idea, in theory at least.
The second function of Dynagroove is to provide what was described as a sophisticated variety of "automatic loudness compensation," which is supposed to correct for the ear's changes in frequency response at different volume levels. When the orchestra plays softly, the Dynagroove computer boosts bass response. As the orchestra plays more loudly, the bass boost is removed and the upper frequencies are boosted. The adjustments are made almost instantaneously, so that a single bass note in the brief pause between two loud trumpet notes is boosted.
The loudness compensation is evidently intended as a means of restoring some illusion of wide dynamic range to a disc with rather narrow dynamic range on it, but if this is its intent (that is, if we got the story straight), there would seem to be some errors in Victor's reasoning, because Dynagroove's action would tend to reduce the illusion of wide dynamic range.
Manipulating frequency response in accordance with the Fletcher-Munson constant-loudness curves could help to give an impression of more dynamic range than is actually present, by introducing the response changes that the ear would normally exhibit with large changes in volume. But this would call for increasing bass during crescendos and reducing it during soft sections (footnote 1). Dynagroove does just the opposite, although it does boost highs during loud passages, which is consistent with the Fletcher-Munson curves.
We were assured that Dynagroove was not predicated on the limitations of the average low-fi phonograph, but the fact that these compressed Dynagroove discs do fare so well on limited-fi phonographs (we tried them on a boom-box and a portable stereo set) makes us wonder about this. Cutting bass during crescendos does put less demand on phonographs of limited tracking ability and limited power capacity, and boosting bass during every quiet section does help to give the impression of more bass than is actually there. But the workings of this ingenious Dynagroove contraption are plainly audible, and disturbing, when the discs are played on a good system.
One thing is clear, though. The "loudness compensation" is an attempt to offset the subjective effects of extreme volume compression, which in turn is predicated on Victor's contention that, and we quote: "Nobody listens at concert-hall volume." Our comments about that are reserved for the editorial in this issue. If we have misunderstood RCA Victor's intent or reasoning anent Dynagroove, we hope to publish their side of the story in the next "Forum" column.
—J. Gordon Holt
Footnote 1: The venerable JGH is actually wrong here, as the ear becomes progressively less sensitive to low frequencies as the level drops, thus requiring more bass, not less. But this doesn't affect the gist of his argument. It should be noted that the amount of pre-equalization used by RCA would be wrong at all playback levels except one, and there was no way of record buyers knowing actually what that level was, or of precisely and repeatably obtaining it in their homes.—JA