by Julian Hirsch
Can Analog Records Match Digital?
Actually, the answer to the question of whether the sound of an LP can match the sound of a CD depends largely on who gives it. It is hardly a secret that a segment of the audio world, small but highly vociferous, is disturbed (sometimes to the point of apoplexy) by the very existence of digital-audio techniques and media. I don't want to get too involved with the more emotional aspects of this question, which may never be settled to everyone's complete satisfaction.
Although I have my own views (dig ital is a giant step forward, with many more pluses than minuses), they are not the point of this discussion, which concerns the results of an interesting listening test.
I don't know how many people have conducted or participated in well-controlled A-B comparisons between digital and analog sound sources originating from the same master recording (a critical requirement unless you are one of those rare individuals who claim to have perfect recall of sounds heard in the distant past). Until recently, I had not had the opportunity to make such a comparison, without which almost no valid conclusions about the sonic limitations of either the digital or analog medium (as distinguished from recording techniques, etc.) can possibly be drawn.
Not long ago, a phono-cartridge manufacturer introduced a new product whose performance presumably represented the current state of the art in analog-record reproduction. Rather than make such a claim directly, the company decided to let its cartridge speak for itself, as it were. A group of audio journalists and reviewers was invited to hear an A-B comparison of the new cartridge and a Compact Disc player. Fortunately, the necessary source material was available in the form of a Sheffield direct-to disc LP record and a Sheffield CD, both derived from the same original performance (the CD having come from an intermediate stage of digital tape recording).
A switching system was constructed that channeled the output of either the phono preamplifier or the CD player to the power-amplifier section of a high-quality integrated amplifier. The levels of the two sources were matched very closely, and every part of the system following the input selector was common to both sources. Signal lights identified the selected pro gram.
It was only slightly surprising to me that the LP and the CD sounded virtually identical-after all, the cartridge manufacturer would hardly have bothered with this demonstration if it had shown his product significantly inferior to the digital disc. I expected the overall frequency response and balance of the two sources to be very much alike, but I thought I'd be able to identify the LP by what I presumed would be a higher noise level. With one minor exception, however, I could not differentiate between the two sources because of noise level. I had also anticipated some difference in clarity as the result of the various forms of tracing distortion, IM distortion, and other nonlinear distortions that are inherent in the analog record-playing process but are essentially absent from digital recordings.
Again, no such effects were audible to me (or, I believe, to the other listeners) in this comparison.
I said that the analog and digital systems sounded virtually identical.
There was a difference, sometimes (but not always) audible depending on the program material. One of the two sources had what I can only describe as a "warmer" sound, which I would guess to be the result of a frequency-response emphasis in the lower midrange or upper bass.
This low-end rise was most apparent on vocal passages, and was very slight. Again, purely as a guess, I would expect a difference in frequency response of less than 1 dB, over an octave or two of the frequency range, to produce such an effect.
The earlier reference to a noise-level difference concerns a couple of very faint "clicks" during the first few seconds of the comparison, which I heard on only one of the sources. This was a "giveaway" to the identity of that source, which was (of course) the LP record. Most LP's exhibit some such noises near their outer diameters, no matter how quiet their surfaces may be elsewhere (and Sheffield records have earned an enviable reputation for quiet surfaces). On the other hand, even the most ardent critics of the CD concede its total silence when the disc is made from a well-recorded digital master tape.
While the "warmer" of the two sources was the LP, I would not characterize it as "better" or "worse" but merely as very slightly different. The reason for the effect is not easy to determine. There are simply too many differences in the processing, manufacturing, and playback stages of analog and digital discs for anyone to be dogmatic about the causes of a slight tonal difference. Even the time-honored comparison against the master tape is impossible here, since the LP was a direct-to-disc recording and had no "master tape," while the digital disc was presumably a bit-for-bit duplicate of the digital master tape.
Of course, the real "winner" of this demonstration was the cartridge, whose sterling qualities could not be open to question after this most impressive performance.
Even so, I have no way of knowing whether other top-quality cartridges could have done as well. I suspect that some could have, but I do not wish to detract in any way from the achievement of this one and the confidence of its developer in its ability.
So, what was proved by this exercise? First, of course, the manufacturer demonstrated that his cartridge is indeed excellent in respect to all the usual performance criteria and perhaps in others less well de fined or understood. Moreover, the listening comparisons showed that when a state-of-the-art analog record is played by a similarly advanced cartridge, the result can be for all practical purposes (except noise levels) indistinguishable from the output of a digital Compact Disc.
It was demonstrated that the various distortions inherent in the analog system, all of which are from ten to a thousand times greater than any of the distortions in the CD play back process, do not seem to be as glaringly obvious as one might expect, even in a careful A-B comparison. And the supposedly deleterious effects of the sharp cutoff filters used in digital recording and play back, including their large phase shifts, are probably no more audible than the many flaws in the LP-record system.
Finally (and, to me, the most important point to remember), these tests proved that the skill of the recording engineer and others involved in the original recording and its duplication are much more important to its ultimate quality than whether it is ultimately produced in analog or digital form. Sheffield LP's are good, and most of them will sound better on any system than the typical mass-produced record. Still, after some years of use, wear, and dust, even they will lose some of their sonic perfection. But a CD should go on forever with no degradation of sound quality.
Also see: THE HIGH END (Jan. 1986)