Cambridge Audio CD1 CD player

Four years after its launch, the CD medium would appear to have come of age, at least in production terms. Annual player manufacture is now big business, and there is hardly a major audio brand without a CD machine to its name—even such analog stalwarts as Audio-Technica and Shure have succumbed.

A huge range of models is apparently available—"apparently" because many machines are repackaged versions of a few basic designs. Prime originators include Yamaha, Philips, Nagaoka, Toshiba, Kyocera, Denon, Hitachi, and Sony, these serving certain specific model requirements as determined by such manufacturers as Audio-Technica, Lux, Sansui, Shure, Mitsubishi, NAD, Micro Seiki, Marantz, Magnavox, Sanyo, and Fisher, to name but a few! Philips' technology, often in the form of complete players, forms the link to Mission, Meridian, CAL, Kinergetics, Sonographe, AR, Proton, B&O, Cambridge, PS Audio, McIntosh, and Revox, as well as the new series of Rotel and Tandberg players.

However extraordinary it may seem, the majority of players sound significantly different from each other; given the concept of an identifiable (but at present unattainable) idea of sonic perfection, each shows its own degree of failure or success in its approach to that ideal.

One model in question, the Cambridge Audio CD1, remains a most useful reference machine; in several respects, it forms the cornerstone of this review, though stories circulate concerning its below-average reliability and an unacceptable failure level in transit—things that should be history, now that it is distributed in the US by Sumiko. My sample dates from early 1986, and has continued to perform well over this extended test period. On the basis of considerable listening tests, if I score this deck at a notional 11.5 points (with the finest analog disc replay currently rating around 15), other CD players currently on sale have sound qualities covering an amazing 3.5 to 12-point range.

The worst-sounding are of music-center, rack-system quality; they may play the discs, but subjectively they have little to do with real hi-fi. Nonetheless, they comfortably beat some of the worst, all-plastic, music-center turntables, though that's not saying very much! These sonic underachievers are not necessarily the cheapest, and there are a few budget CD machines offering pretty good sound for the money, especially those of Philips derivation. On a statistical basis, Philips-based designs have consistently achieved the best review points.

There is a fascinating contrast between older Philips machines: 14-bit, dual DAC chip, 4x oversampled, noise-shaped and digitally filtered designs; and the newer 16-bit, dual DAC/one chip, 4x oversampled and digitally filtered units. In the hands of manufacturers such as Meridian, the older designs typically achieve a fine 15.5-bit resolution—ie, the least-significant bit has a 50% chance of being wrong—while a surprising number of "new tech" Philips 16-bit players have been delivered with substandard 14-bit resolution, only the most costly examples exceeding 15-bit precision.

One can easily make too much of the player technology and of any specific parameter associated with the typically excellent lab results generally obtainable from CD players. In terms of sound quality, factors such as linearity, bit count, or ultimate channel separation appear to be less influential than structural integrity, disc damping and clamping, error control, transport decoupling, the segregation of analog and digital sections, integrity of power supplies and internal grounding, as well as dual-mono construction, and, not least, component quality—all of which seem to have a strong influence on sound quality (although not all are of equal importance).

Equally relevant is the implementation of the circuits used for analog filtering and final output stages. The judicious use of a specific grade of decoupling capacitor at a unique circuit position can sometimes have a remarkable influence on sound quality.

It is worth noting that almost none of the above remedial factors have any effect on the measured lab performance; indeed, some of the "improvements" verified sonically have resulted in poorer test results: high-level linearity, for example, with a consequent increase in distortion at full modulation. As with the best preamplifiers, such contradictions only serve to illustrate our ignorance of the relationship between the objective and subjective domains of audio engineering.

Alert readers will have noticed an alarm bell ringing earlier in this introduction, namely "top analog at 15 points"—three ahead of the best current CD replay at 12. For all the marketing pressure of the digital bandwagon, the fact remains that, at present, analog vinyl has the edge. There are only too many reminders. Try Paul Simon's Graceland, an analog recording which sounds noticeably poorer in its digital transcription replayed via CD.

I recently had temporary use of a Goldmund Studio turntable fitted with the T3 arm and a Koetsu Signature Red. While I was not wholly convinced by the ultimate precision of the high treble provided by this arm, the overall performance of the vinyl replay—via a Cello "Premium" Suite preamplifier, Krell KMA100 II, and biwired (and custom modified) Celestion SL600 speakers—was something special. Good CD, with appropriately ambient recordings, can sound dimensionally spacious and is capable of invoking pretty good depth impressions, but the "small" Goldmund's ability to build a deep, focused soundstage, fired with subtle musical detail and involving dynamics, went far beyond the current top CD standard.

It is the responsibility of the CD-player designer and those involved in the CD sector of the recording business to find an answer to this discrepancy if CD is not to remain "just another source," as AJ van den Hul recently put it (Vol.9 No.8, pp.135–142).

Cambridge Audio CD1 $2800
This substantial-looking two-box machine comprises a lower deck containing the decoder and analog filter sections, these separated and screened from the larger upper deck that carries control, transport, and power-supply sections. Finished in typical Cambridge style, with dark teak end-panels in solid wood and an overall matte-black finish, this player offers two pairs of output terminals, one fixed, the other passively variable via a moderate-impedance potentiometer.

The player transport is based on a modified version of the Philips CD104/Magnavox 1040 that incorporates a diecast metal chassis. A drawer-loader, this player may be programmed for up to 20 tracks, albeit laboriously, while the multipurpose display is a small, green, fluorescent type that needs looking at closely if it is to be read easily. High-quality pushbutton switches are used for control, forming a dominant row across much of the width of the machine. A secondary set of three switches may be used singly or in connection to provide a variety of filter functions, numbered 1 to 6, 1 corresponding to the standard flat response.

The player uses Philips' digitally filtered, 14-bit, 4x oversampled system, but with a difference: three DACs are used per channel, two in parallel and one for ranging. Dither is added to improve the resolution, rounding the result up to 16-bit overall. While the analog circuitry is based on 5534-type op-amp ICs, this section is carefully built with good power supplies, van den Hul mono filament cable, and Wondercaps. Care has been taken over the mechanical construction—lead metal bars damp the CD transport and provide the suspension with a low resonant frequency, said to be tuned to 1Hz (footnote 1).

Sound Quality
This player has stood the test of time; it continues to give the best-focused, most stable, and widest stereo images I have yet heard from CD. The sound was dynamic, with a fast, strong, articulate bass that carried a surprising punch on the right materials. Essentially well balanced, the sound of the Cambridge featured a consistently high level of clarity and instrumental separation, particularly good when it came to treble timbre and the differentiation of tonal color. Stereo depth was very good, only (just!) beaten by the CAL Tempest, though its stereo stage was a little forward.

The sound could be faulted slightly, due to a hint of coarseness in the mid and treble—a touch "solid-state," if you like. Nevertheless, it was clearly more elegant than most of the competition, even in this respect. Sonically, the CD1 had a lively bounce, with a good sense of involvement in the music.

The variable output was inferior, but proved useful for driving a power amplifier direct via short, high-quality cables. We preferred filter 1 for most listening work.

The Cambridge Audio CD1 is a reference-standard, true 16-bit resolution, 4x oversampled player with as yet unrivalled dynamics, stage width, and precision. Suited (just) to rock more than to classical material, the sound quality was ranked very close to that of the other tested reference, the CAL Tempest—personally, I would not like to choose between them.

As regards features and operation, the CD1 is somewhat primitive, but nonetheless a worthy contender. At the time of writing, rumors of an improved Cambridge CD1 were circulating, as well as the possibility of smaller versions that would come close to replicating the performance of the existing machine at much less cost. One story even hinted at a prototype Cambridge with 16x oversampling, which will place the center of the first alias image at 0.352MHz, allowing very simple passive filtering to be used well away from the audio band, with a consequent improvement in audible transparency. Conversely, very wide bandwidth circuitry will be required to avoid slew and high-frequency intermodulation problems.

Footnote 1: The CD1 was designed by Stan Curtis.—Ed.
Cambridge Audio
US distributor: Audio Plus Services
156 Lawrence Paquette Industrial Drive
Champlain, NY 12919
(800) 663-9352