Not only were noise levels low, but so was distortion. Even into a punishing load of 600 ohms, the spectrum of a full-scale 61Hz sinewave (fig.6) showed no harmonic products. (The spikes between 600Hz and 700Hz in this graph appear so often in Audio Precision FFT-derived plots that I suspect they are mathematical artifacts.) With the high-frequency intermodulation torture test (fig.7), the 1kHz difference component does poke its head above the noise floor, but at -96dB it is still at an inconsequential level, even into 600 ohms. The Sondek must have a bomb-proof output stage.

Fig.6 Linn Sondek CD12, spectrum, DC-1kHz, 61Hz at 0dBFS, 600-ohm load (linear frequency scale, 20dB/vertical div.).

Fig.7 Linn Sondek CD12, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 0dBFS, 600-ohm load (linear frequency scale, 20dB/vertical div.).

I used the Miller Audio Research Analyzer, which is based on a National Instruments "virtual instrument" DSP PC card, to examine the Sondek's jitter. The resulting high-resolution spectral analysis is shown in fig.8. The peak-peak jitter level was among the lowest I have measured, at 136.6 picoseconds, with a negligible time-base error of just +3ppm. The only significant jitter components that can be seen in fig.8 are the data-related spectral lines at multiples of +/-229Hz (indicated with red numeric markers). The highest-level sidebands are at +/-15.6Hz (marked with a purple "1") and are almost indistinguishable from the central peak. I have seen these sidebands before in players or processors using the PMD100 digital filter.

Fig.8 Linn Sondek CD12, high-resolution jitter spectrum of analog output signal. (11kHz at -6dBFS with LSB toggled at 229Hz.) Center frequency of trace, 11kHz; frequency range, +/-3.5kHz.

Overall, this set of measurements is about as good as I have seen from a CD player. The Sondek CD12 obviously has been the beneficiary of some excellent Scottish engineering!
—John Atkinson