CD Player/Transport Reviews

The Sony CDP-X779ES had an output of 2.49V (L) and 2.52V (R), unbalanced when decoding a 1kHz, 0dB (full-scale) sinewave. The corresponding balanced outputs were 2.1V (L and R). Its output impedance, being well below 1 ohm, was actually too low to measure on our Audio Precision test set in either the balanced or unbalanced (fixed or variable) modes. The Sony's DC offset was 0.1mV left and 0.2mV right. The '779 was non-inverting from the unbalanced outputs, with a positive-going impulse test signal reproduced as positive at its outputs. I verified that the balanced output is configured with pin 3 positive, as specified (the opposite of the AES/ANSI standard). Its frequency response (fig.1) was almost ruler-flat, with a barely perceptible rise above 10kHz. The figure shows the unbalanced output; the balanced output was dead flat to 20kHz. The de-emphasis error, also shown in fig.1 (bottom curve), indicates that those few discs which use pre-emphasis will be played back with the spectral response inherent in the recording.

The crosstalk shown in fig.2 is very low even from the unbalanced outputs, and outstanding from the balanced. With the latter, separation from right to left is slightly better than the reverse, but neither is significant in any way. Fig.3 shows the spectral analysis of the Sony's decoding of a track of "digital silence" (all data words zero) from the balanced output. Note the very low overall noise—dropping below our standard –130dB graph floor below 1kHz—and only a modest rise at the 60Hz line frequency, still below any conceivably audible level. The steep rise in noise above 20kHz is typical of low-bit noise shaping and not a matter of concern.

Using the same type of spectral analysis, fig.4 shows the spectrum of a decoded –90dB, 1kHz dithered sinewave signal (unbalanced output shown, the balanced was marginally better at 60Hz). Only minor artifacts are visible above and below the test-signal frequency, with the latter showing excellent linearity. Fig.5 shows the fade-to-noise with dither, again indicating outstanding linearity down to –110dB. Only the left-channel balanced output is shown; the right channel and the unbalanced outputs were virtually identical.

Noise modulation as a function of signal level vs frequency is plotted in fig.6. (The test signal used for this test was from Stereophile's Test CD 2, which includes a full description of this test.) The results shown here are for six different signal levels decreasing from –50dB to –100dB. The more tightly clustered the results, the better; the Sony's grouping is first-rate. Many of the curves in some of our previous measurements of this characteristic in other players showed a rise at the lowest frequencies, leading me to suspect a possible measurement artifact. The Sony, however, except for the –50dB curve (which increases below 600Hz to –110dB) and the –60dB and –70dB curves (showing single spikes at 420Hz and 310Hz, respectively) was very well behaved down to the lowest frequencies, indicating that the low-frequency increases measured on other machines were most probably due to the players themselves.

Fig.7 shows the results of the Sony decoding a 1kHz, undithered sinewave at –90.31dB. The desired stairstep response is very clear—superior to that which we have observed in the majority of players and processors tested to date. Feeding a full-scale combined 19kHz+20kHz signal into the Sony and performing an FFT analysis of the output resulted in the plot shown in fig.8. The artifacts are very low, with the highest—the 1kHz difference frequency—over 90dB down.

Using a full-scale (0dBFS) 1kHz squarewave to drive the Sony, we got a result typical of players using linear-phase digital filters, but with no clipping of the ringing on the top and bottom of the squarewave noticeable (fig.9). Jitter tests were not performed with the Meitner analyzer on the X779ES because its digital circuitry is not accessible with the player. I did get a peek inside this large, heavy machine, however, which on whole appears well made, though the reasonably neat layout does include some not particularly tidy wiring bundles connecting the several circuit boards.

Altogether, the measured performance of the CDP-X779ES was outstanding. Nothing in the results would indicate anything but first-rate performance across the board.—Thomas J. Norton