Examining the DC-101's analog distortion proved fruitless, as any harmonics were below the noise floor of the Audio Precision System One measurement system. Fig.7, for example, shows an FFT-derived spectrum as the processor decoded PCM data representing a full-scale 50Hz tone into a punishing 600 ohm load. Yes, harmonic components—the 6th, 7th, 8th, 15th, 17th, and 18th—can be seen above the -120dB level, but I suspect these are actually due to the AP's lack of mathematical dynamic range. The story was similar regarding intermodulation distortion. Even with the Accuphase driving a torture-test signal of combined 19kHz and 20kHz tones into 600 ohms (fig.8), the 1kHz difference component was still no higher than -97dB (0.0014%).

Fig.7 Accuphase DC-101, CD data, spectrum of 50Hz sinewave, DC-1kHz, at 0dBFS into 600 ohms (linear frequency scale).

Fig.8 Accuphase DC-101, CD data, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 0dBFS into 600 ohms (linear frequency scale).

To examine the effect of word-clock jitter, I played in the DP-100 transport (connected to the DC-101 with the HS-Link data cable used for SACD replay) a CD-R carrying a high-level sinewave at one quarter the sample rate (11.025kHz), the LSB toggling on and off at 229Hz. The DC-101's unbalanced analog output was fed to a National Instruments DSP-2200 PC card, with the host PC running the Miller Audio Research Jitter Analyzer software. This averages sixty-four 32,768-point FFTs and searches the resultant high-resolution spectrum for symmetrical sidebands on either side of the 11.025kHz tone. Fig.9 shows the result: The absolute level of jitter was very low, at 168.5 picoseconds peak-peak, and consisted almost entirely of data-related components, indicated here with red numeric markers. However, quite a few low-level noise spikes can be seen (blue markers); these could be associated with the HS-link, which uses a network-type twisted-pair topology.

Fig.9 Accuphase DP-100/DC-101 joined by HS-Link, CD data, 44.1kHz sampling, high-resolution jitter spectrum of analog output signal (11.025kHz at -6dBFS with LSB toggled at 229Hz). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

I therefore repeated the test driving the DC-101 from a PS Audio Lambda CD transport via a conventional S/PDIF cable. The jitter (not shown) rose to 176ps, mainly due to the introduction of pairs of sidebands at ±60Hz (the AC supply frequency) and ±90Hz (no idea), but the noise floor otherwise remained the same.

I then used the DC-101 to decode SACD data supplied by the DP-100 transport. Fig.10 shows the frequency response at -3dBFS, measured using the "provisional" Sony test SACD. The audioband response is flat, with then an HF rolloff reaching -3dB at 50kHz and -15dB at 75kHz.

Fig.10 Accuphase DC-101, SACD data, frequency response at -3dBFS (right channel dashed, 2dB/vertical div.).

Fig.11 shows wideband 1/3-octave spectral analyses taken of the DC-101's analog outputs while the DP-100 supplied SACD data representing 1kHz sinewaves at -100dBFS and -120dBFS. (Both tones are below the cutoff point of undithered CD data.) Each tone's level is correct, and although spectral content can be seen at 60Hz and 120Hz, as I explained in the measurements accompanying Jonathan's November 1999 review of the Sony SCD-1 SACD player, I suspect this content is actually on the test disc.

Fig.11 Accuphase DC-101, SACD data, 1/3-octave spectrum of dithered 1kHz tones at -100dBFS and -120dBFS, with noise and spuriae (right channel dashed).