Sony CDP-101 Compact Disc Player JGH Responds to Doug Sax

J. Gordon Holt then offered the following controversial thoughts in response to an article (linked here) from Doug Sax of Sheffield Labs and The Mastering Lab in December 1983, Vol.6 No.5:

Run Right Out
I have never before done this, but I am going to recommend a product to all of our readers who can afford it. I am referring to the Compact Disc player.

One of the few things that our entire staff has ever agreed about is that the sound obtainable from digital audio can be better than the best that available from analog sources, particularly home analog sources. (Larry Archibald is undecided about this; J. Gordon Holt, Dick Olsher, and Bill Sommerwerck are sure.) What makes this approximately $1000 purchase so worthwhile is that a good 95% of audiophiles are not getting anything even approaching the capability of analog from their systems, and probably never will. I've belabored this point before, and don't feel the need to do so further at this time. Suffice it to say that no two "experts" have ever agreed as to the best cartridge, tonearm, turntable and/or preamplifier. That may say something about the experts, but it says more about the total anarchy which exists in the analog front-end (footnote 1) field today. And while accuracy (to the recording) is a watchword among audiophiles, euphonics (does it sound good?) are what determine how an audiophile feels about reproduced sound. Few seem willing to question how it can be that six phono units, all sounding "musical" but different in various ways, can all claim to be "accurate."

I know, I know. I've read the hysterical tirades in The Absolute Sound, and I've read Doug Sax's impassioned letter to 140 high-end manufacturers as well as his Billboard article. But there are reasons to doubt the impartiality of both with respect to digital audio. TAS painted themselves into a corner before they had a chance to actually try a digital recording system (they still haven't as far as we know), so their continuing anti-CD stand could well be, among other things, a face-saving device. Sax's livelihood is the cutting of analog discs. This is not to say that I doubt the sincerity or integrity of either; just that both have good reasons to see CD lie down and die, something it's not likely to do at this point.

Sheffield may be adamantly negative about digital right now, but other perfectionist-oriented record labels are coming around to digital. Mobile Fidelity is now releasing digital cassettes in VHS format, and Direct-to-Tape Recordings (in Haddon Heights, NJ) is routinely making all of their material available on the buyer's choice of VHS or Beta digital. It's my feeling, though, that digital audio on videocassette is just a high-priced stopgap. I don't think it has a hope of competing with the CD, simply because a blank cassette costs more than a finished CD and PCM from videocassettes often suffers from playback machine incompatibility problems. A CD recording now costs about the same as most audiophile analog releases, and offers the potential for much better sound. And CD production prices are already low enough that, once the supply starts to catch up with the demand, prices should begin to come down. They may not go below $10 each for some time (if ever), but even that is not much higher than you pay for a regularly priced LP. CD discounting may even start to undercut that $10 figure.

Most audiophiles, who presumably have no vested interest one way or the other, have been ecstatic about the CD sound; a sizable number, however, have not been impressed. There are, in fact, some seemingly legitimate theoretical criticisms levelled at CD and digital recording, to wit: the sampling rate is too low; the bit rate is too low; the anti-aliasing filters cause terrible phase shift. And so on and so on and so on. Frankly, I do not give a tinker's dam about objections based on theoretical desiderata. What I do care about is how something sounds, and when something sounds as good as CDs can sound, then we should be questioning the justification of those criticisms instead of trying to convince ourselves that the sound is horrible when in fact it is not.

But what about all those bad things that some claim to be hearing from digital? It has been asserted that digital's finite noise floor is an impenetrable barrier, below which the system can reproduce no signal. That should kill ambience, shouldn't it? But it doesn't, because a process called dithering (footnote 2) is used to extend that floor downwards by about 15dB, yielding an effective dynamic range (not to be confused with signal/noise ratio) of around 105dB. This is almost identical to the best analog tape recorders with noise reduction (on which it's also possible to hear signal below the noise floor). Interestingly enough, as you might guess from the above statement, dynamic range should not be seen as digital's big advantage: already the full range of analog is rarely utilized, not to mention digital. Checking out what's available on Compact Discs, we found the dynamic range not even as great as some analog discs (Vol.5 No.10, p.10). This does not include measurements on the new Telarc discs, which we're just now attempting to make (and which promise to be really something).



Footnote 1: In Stereophile parlance, a "front end" is the whole schmier that delivers signal to the power amplifier, and includes the cartridge, arm, turntable and preamp.

Footnote 2: The concept and importance of dither is explained quite fully by Martin Colloms in Hi-Fi News & Record Review, August 1983.

X