Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting Departments | Features | ADs | Equipment | Music/Recordings | History |
Manufacturer's Specifications Frequency Response: 2 Hz to 20 kHz, ±0.3 dB. S/N: 106 dB (EIAJ). Dynamic Range: 97 dB (EIAJ). THD: 0 0025% at 1 kHz (EIAJ). Separation: 100 dB at 1 kHz (EIAJ). Wow and Flutter: Less than 0.001% (EIAJ). Number of Programmable Selections: 20. Output Level: Fixed, 2.0 V; variable, 0 to 2.0 V; headphone, 28 mW into 32 ohms; digital output, 0.5 V peak to peak. Power Requirements: 120 V a.c., 60 Hz. 16 watts. Dimensions: 16 15/16 in. W x 4 1/8 in. H x 15 9/16 in. D (43 cm x 12.3 cm x 39.5 cm). Weight: 29.7 lbs. (13.5 kg). Price: $1,500. Company Address: Sony Dr., Park Ridge, N.J. 07656. When I measured and listened to Sony's CDP-650ESD two years ago, I thought that the science of digital audio disc reproduction had reached its ultimate level. I should have known that when the subject is sound reproduction, there's always room for improvement-however small in audible terms. The company's latest effort in its ongoing crusade to deliver the best digital sound possible is the not inexpensive CDP-705ESD Compact Disc player. I suspected that this unit was something out of the ordinary when I lifted it onto the test bench: It is probably the heaviest CD player I've tested. While weight is certainly no measure of how an audio component performs, some of the CDP-705ESD's heftiness does, indeed, impact on performance. The weight is largely due to a new chassis material, a specially developed calcium carbonate substance rein forced with glass fibers to make it more rigid, according to Sony, than conventional metal frames. Sony also states that this chassis has been acoustically tuned to a low system resonance so that external or internal vibration transfer is reduced. Additionally, the chassis is nonmagnetic. The player's entire bottom panel, back panel, and center partition are copper plated, then finished in black for cosmetic symmetry. Mechanical vibration is limited by separating the main chassis from the base with rubber bushings. The internal resonance of the base is suppressed by vibration- absorbing material that has been applied at key structural locations. This player also has large insulator feet to help keep mechanical and airborne vibration from affecting performance. All this is done to reduce vibration which, if excessive, might induce laser reading errors or loss of data. Excessive vibration also forces the system to interpolate sample readings, and excessive interpolation results in distortion. Thus, anything that can be done to reduce the need for such interpolation will likewise reduce distortion. Sony's compact Linear Motor laser-transport mechanism, with its combination of fast access speed and precise cueing, has been carried over into this design. So have such earlier refinements as the poly-resin ceramic material used to damp any vibrations that might otherwise occur as the laser-optical assembly moves rapidly from one point to another. Electrical interference is reduced through the use of highly stabilized power supplies and the inclusion of copper bus bars in various circuits to avoid current loops. High-speed switching diodes are used as supply rectifiers, and RC filters are inserted in the secondary side of the power transformer to resist external noise and to suppress hum that might be transmitted to other audio components. Separate digital and analog power supplies and sub-regulators are also used to help minimize electrical interference be tween the two sections. In terms of electronic circuitry, several refinements are incorporated in the CDP-705ESD. Instead of two-times oversampling, Sony has elected to use four-times oversampling along with true 16-bit, dual-D/A conversion. While providing no additional music information, this technique has the ad vantage of decoding at a frequency four times higher than the standard 44.1-kHz sampling rate. The result is that unwanted noise sidebands are shifted to a higher frequency range (centered about 176.4 kHz). This allows a gentler filter to be used in the final analog stage to eliminate the sampling frequency. The result is phase accuracy throughout the audible spectrum as well as more uniform frequency response near the treble cutoff point. The digital filter circuits in one of the new LSI chips which Sony has used in the CDP-705ESD, in tandem with new high-speed D/A converters, hold band-pass ripple to less than 0.001 dB while providing more than 80 dB of stop-band rejection at any frequency above 25.7 kHz. Sony has also developed what they call an advanced Error Prediction Logic tracking servo circuit to control the position of the laser pickup. This circuit monitors the pick up's position 8,000 times per second. When an error is first detected, the circuit memorizes its parameters. On the next pass or revolution of the disc, the same signal parameters are compared by a microprocessor to the information memorized on the first pass and then noted. The circuit next determines the probable location and duration of the error before the third disc rotation, after which it directs the servo circuit to respond with the exact amount of current necessary to maintain proper tracking. This circuit divides each disc rotation into more than 1,065 discrete intervals, monitoring about once every 0.34°. Circuit features carried over from earlier models in Sony's ES series include the Envelope Differential Servo Detection system, which detects any dropouts of digital data as soon as they occur, allowing for error correction before tracking error results. The same circuit "freezes" the laser pickup in position when no digital data is present, providing faster recovery from gross disc defects and more uniform tracking. Sony's Unilinear Converter is also present in this model, serving to reduce spurious, out-of-band "beats" caused by the interaction of multiple clock rates such as those found in other oversampling filter designs. Optical coupling has again been used between the digital filter stage and the demultiplexer and delay stages to isolate the audio stages from the digital circuits that precede them. For all its internal refinements, the CDP-705ESD offers about the same operating and convenience features that have been available on the company's previous ES-series players and even on their lower-cost models. Direct track access (using the numbered keys on the front panel or on the supplied remote control) is available, as is direct index access. Up to 20 selections can be programmed for play in any order. The various modes of repeat play (including repeat of a defined musical section) are all here, as is the popular "shuffle play" Sony introduced some time ago. This mode randomizes the order in which the tracks on any CD will be played. Fixed and variable outputs as well as a headphone output are provided, and the variable output level can be controlled using the remote. A digital output jack is also included. This is nothing new; several makers have offered subcode and direct digital output terminals on earlier CD players. However, Sony tells me that this redesigned digital output stage fully conforms to industry format standards and also includes a new digital "latch" for correction of any time-base errors caused by jitter. When the digital port is used, the player's analog stage is completely bypassed. Furthermore, one will be able to use the single digital output for graphic subcode data as well as for direct transfer of digital audio data. Previously, separate ports were needed for these two functions. Control Layout The front-panel layout of the CDP-705ESD is reminiscent of earlier Sony ES players, though with some rearrangement of buttons and switches. The "Power" pushbutton, a head phone jack, and a three-position "Timer" switch (for use with an external clock timer) are at the panel's left end. The smooth-acting but ruggedly constructed disc tray is positioned above seven small pushbuttons which handle the various repeat-play modes, time/program display options, the "Shuffle" mode mentioned above, "Auto Pause" (which, after each selection plays, cues up the next track and puts the unit in pause mode), and "Auto Space" (which inserts a pause of approximately 3 S between selections). A large display area to the right of the disc tray shows track and time information and also incorporates the "track calendar" introduced by Sony a few years ago. This is a grid with 20 numbers for as many tracks; when a disc is inserted in the tray, numerals corresponding to the tracks available on the disc are illuminated. As tracks are played, the track numbers are extinguished. When random-access programming is used, only those numerals included in the program illuminate. Just below the display area are six small, light-touch buttons used for skipping ahead or backwards from track to track, for advancing to the next or previous index points within a track, and for fast forwarding (or reversing) the laser pickup while monitoring the output at reduced volume levels. Below these six keys are four large buttons, controlling eject, play, pause, and stop. One minor feature of the "Index" buttons eluded me for a while, since it is not mentioned in the owner's manual. When I tried to move backwards from one index point to another, no matter how many times I pushed the reverse index button, the best I could do was to get the laser to return to the start of the current index point. It was only when I inadvertently kept the button depressed for a few seconds, rather than for just a touch, that the laser assembly did, finally, move back from index point 3 to index point 2 of the track I was playing. At the upper right are the number buttons used for directly accessing tracks and for programming the machine. "Check" and "Clear" buttons, used during programming, are also found here, as is a "+ 10" button for programming track numbers higher than 20 (if you happen to own a CD with that many selections on it). Three light-colored buttons near the lower right corner of the panel are used to select programmed play, single-track play, or continuous play of an entire disc. A tiny rotary level control in the lower right corner completes the panel layout. When the remote control is used to alter output levels at the variable output jacks, this tiny knob actually rotates, since the gain is being altered by a motorized precision potentiometer. The rear panel is equipped with pairs of fixed and vari able analog output jacks as well as with the single digital output jack mentioned earlier. A slide switch above the digital jack must be in the "On" position for the digital code's signal to appear at this jack. With the switch on, regular analog audio output is not available at either set of line output jacks or the headphone jack. Measurements Figure 1 shows frequency response of each channel as the response sweep track of my new EIA test disc was played on the CDP-705ESD. Maximum deviation from absolutely flat response within the audio range was no more than +0.2 dB and -0.6 dB, with the maximum negative deviation occurring only at 20 kHz. Harmonic distortion at 0-dB (maximum) recorded level was by far the lowest I have measured for any CD player. At most frequencies, THD was actually less than 0.002%. (For all I know, since the "floor" of my THD measuring equipment is not much lower than 0.002%, it may even have been still less!) A plot of THD versus frequency is shown in Fig. 2. The dotted line corresponds to the readings taken with the 20-kHz low-pass filter removed from the measurement path. It indicates that the higher readings (never higher than 0.5%, in any case) result from out-of-band signals which are not necessarily harmonically related to the test frequency.
Figures 3A and 3B show, respectively, unweighted and A-weighted signal-to-noise ratio. Overall unweighted S/N measured 95.0 dB, and A-weighted S/N was 99.7 dB. The benefit of Sony's Unilinear Converter system and the use of a single master clock to synchronize all digital functions is clearly seen in Fig. 4, in which the 20-kHz output of the test disc was analyzed using a spectrum analyzer sweeping from 0 Hz to 50 kHz. One can see the usual 24.1-kHz out-of-band "beat," caused by the interaction between the 44.1-kHz sampling rate and the 20-kHz test signal. The beat is, in fact, rather high in amplitude because of the gentle analog filtering used in this player. But you will notice that there are absolutely no other spurious outputs visible, either above or below the test frequency (represented by the taller of the two spikes in the 'scope photo). Most players, even those that use oversampling and digital filtering, exhibit additional spurious products both outside and within the audible frequency range.
Stereo separation was also the highest I have obtained for any CD player. At low and mid-frequencies, separation measured a full 100 dB, decreasing only very slightly to 95 dB at 10 kHz and to 92.0 dB at 16 kHz. Figure 5 shows separation versus frequency from left to right and from right to left; only a single curve can be seen because separation was identical in both directions. Dynamic range, measured in accordance with the prescribed EIAJ and EIA methods, was an impressive 107 dB, and wow and flutter was too low to be measured with my test equipment. In the past, I have always run into trouble when trying to measure output linearity below around-70 or-80 dB. I assumed that this was caused by limitations in my test setup and was surprised to find that this is not the case. The CDP-705ESD exhibited near-perfect linearity (within 0.1 dB) down to 80 dB below maximum recorded level and was off by less than 1.0 dB at-90 dB! As far as I have been able to determine, very few commercially pressed CDs employ pre-emphasis for additional improvements in S/N and dynamic range, but for those that do, it's important that a player exhibit proper de-emphasis characteristics. This player provided perfect de-emphasis when presented with a pre-emphasized test signal; it was the first player I've tested that didn't deviate by so much as 0.1 dB in this measurement. CCIF (twin-tone) intermodulation distortion was only 0.001% at 0-dB recorded level, increasing very slightly to 0.0015% at-10 dB. Output level for 0 dB, whether measured at the variable output jacks with the gain control set to maximum or at the fixed output jacks, was 2.10 V, with no difference in level between channels. Figure 6 is a 'scope photo of a 1-kHz square wave as reproduced by this player. If the ripple along the horizontal portions of the display seems a bit lower to you than it has in previous reports, all I can say is that I got the same impression. It suggests that this unit's phase accuracy is some what better than that observed in other players. The reproduced unit pulse shown in Fig. 7, besides confirming that oversampling and digital filtration have been used here, also tells me that the CDP-705ESD does not invert waveform polarity. Finally, Fig. 8 shows that dual D/A converters are being used and that time correction has been applied so that there is no phase shift between the 20-kHz signal reproduced from the left output compared with the same signal reproduced from the right. As you might expect, the CDP-705ESD tracked my special "defects" disc perfectly. It also did a remarkable job of tracking several discs that I was ready to discard because of scratches and imperfections which led to sustained mis-tracking and muting on other CD players. I had seen some pretty dramatic demonstrations of the effectiveness of Sony's Envelope Differential Servo Detection system before, but those had been given by people at Sony. It's quite another matter when you experience the incredible tracking ability of this system on your own turf. It is nothing short of remarkable. Use and Listening Tests I can't honestly say that I detected any major audible difference in sound quality between the CDP-705ESD and my reference CD player, the Sony CDP-650ESD, though with certain discs I detected a somewhat more pleasant high end when using the new model. I can attest to the fact that access time has actually been reduced even further, to less than 1 S from tracks near the center of a disc to those near its outer rim. I can also state, with a fair degree of confidence, that tracking ability and stability are better on the CDP-705ESD than on the reference unit. Which of the many mechanical and electronic improvements is responsible for this improved tracking I can't say, but I do know that I was able to tap the top and sides of the CDP-705ESD a lot harder than I can on the CDP-650ESD before there was any evidence of mistracking or momentary muting. One of the discs I used to audition this player was a Denon recording of a Mozart clarinet quintet that I recently bought in Japan. The sound of a clarinet can be somewhat grating if the delicate relationships of all its overtones are in any way altered or upset. The sound of the clarinet solo on the Denon disc was velvety-smooth and warm. Above all, it was totally devoid of the harshness I had heard when playing this recording on a low-cost player. A second recently acquired disc that sounded particularly fine on this player was an all-digital Philips recording of Elgar's "Cello Concerto in E Minor" (Op. 85) that also contains his well known "Enigma Variations." Again, the warmth and richness of the cello, which has been so difficult to reproduce properly, was immediately evident when I played this disc on the CDP-705ESD. I could cite other discs that sounded better on this machine than they had on competing units, but if you can afford this state-of-the-art CD player, you'd probably be more thoroughly convinced by listening for yourself than by taking my word for it. -Leonard Feldman (Audio magazine, Jul. 1987) Also see: Sony CDP-701 ES Compact Disc Player (Sept. 1983) Sony CDP-707ESD Compact Disc Player (Equip. Profile, Apr. 1988) Sony CDP-X77ES CD Player (Jun. 1990) Sony CDP-620ES CD Player (Sept. 1985) Sony CDP-610ES CD Player (Nov. 1984) Sony CDP-650ESD Compact Disc Player (July 1985) = = = = |