Rega Planet CD player Measurements
The Planet's maximum output level was 0.2dB above the CD standard at 2.05V. This might be audible in comparisons, but I wouldn't lose sleep over it. The analog output was absolute-polarity correct, and was sourced from 930 ohms (the measured figure was a little higher at 20Hz, a little lower at 20kHz). The Planet coped quite well with disc data dropouts. It negotiated track 30 on the Pierre Verany test disc but started to skip on track 31, which includes a 1mm dropout. The data clock was set a little high in frequency, an 11.025kHz tone playing with a +182ppm (0.0182%) error.
The Rega's frequency response is shown in fig.1: The channels match to within 0.05dB, and although there is the merest sniff of a response rise above 10kHz, this is not nearly enough to correlate with my impression that the player's highs were slightly tilted-up. The de-emphasis was effectively perfect. The channel separation was adequate, at better than 95dB over much of the audioband. But as can be seen from fig.2, there was some power-supply noise evident, as well as a very strange crosstalk artifact at 850Hz in both channels.
Fig.1 Rega Planet, frequency response (right channel dashed, 0.5dB/vertical div.).
Fig.2 Rega Planet, channel separation (10dB/vertical div.).
The power-supply spuriae can also be seen in the spectral analysis of a dithered 1kHz tone at -90dB (fig.3). But even the highest in level, at the full-wave switching frequency of 120Hz, is barely above -100dBFS; you shouldn't make too much of this. I couldn't hear hum even with my ear pressed against the Thiel CS6's woofer and the volume control turned up (don't try this at home, kids—I am a trained professional). There is a suspicion of a 2kHz component present in both channels, and the slight bulge of the 1kHz peak at 800Hz might have something to do with the crosstalk artifact. But again, not too much should probably be made of this.
Fig.3 Rega Planet, spectrum of dithered 1kHz tone at -90.31dBFS, with noise and spuriae (16-bit data, 1/3-octave analysis, right channel dashed).
Extending the analysis bandwidth to 200kHz and driving the player with data representing "digital black" gave the spectrum shown in fig.4. Again, the 120Hz and 240Hz power-supply artifacts can be seen, as can the peak at 2kHz, but the noise floor is otherwise both low in level and "white" in nature, suggesting that the Planet uses a multi-bit DAC of some kind.
Fig.4 Rega Planet, spectrum of digital silence, with noise and spuriae (16-bit data, 1/3-octave analysis, right channel dashed).
Fig.5 shows the left channel's linearity error, plotted against absolute level. The curve is smooth, with 1dB or so of negative error evident down to -100dBFS. This is good performance, confirmed by the shape of an undithered 1kHz sinewave at -90.31dBFS (fig.6). The discrete voltage steps can be made out, though the trace is overlaid by some low-frequency noise, presumably due to the power-supply breakthrough.
Fig.5 Rega Planet, left channel, departure from linearity (2dB/vertical div.).
Fig.6 Rega Planet, waveform of undithered 1kHz sinewave at -90.31dBFS (16-bit data).
Fig.7 is a spectrum of the player's output while it decodes data representing two high-frequency tones, at 19kHz and 20kHz, each at -6dBFS. The amplitude of the combined waveform just reaches 0dBFS, representing a torture test for a player's analog output stage. The Rega performed pretty well on this test, the 1kHz difference component and the 18kHz and 21kHz second-order components all remaining below -90dB.
Fig.7 Rega Planet, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 0dBFS (linear frequency scale, 20dB/vertical div.).
Finally, fig.8 is a high-resolution spectral analysis of the Rega's noise floor (made with the Miller Audio Research jitter analyzer) as it plays a CD-R encoded with a special test signal intended to reveal the effects of jitter: an 11.025kHz tone at -6BFS combined with the LSB being toggled at a 229Hz rate. The grayed-out trace, supplied for reference, is the noise floor of the Myryad MC 100 player that I reviewed last month. The Rega's jitter measured 382 picoseconds peak-peak compared with the Myryad's 166.7ps under the same conditions. However, while the Myryad's noise floor featured the strange rise around the central, 11kHz peak, which I conjectured might be related to its rather "polite" presentation, the Rega's jitter signature consists of discrete tones.
Fig.8 Rega Planet, high-resolution jitter spectrum of analog output signal (11kHz at -10dBFS with LSB toggled at 229Hz). (Center frequency of trace, 11kHz; frequency range, ±3.5kHz.) Grayed-out spectrum is that of the Myryad MC 100.
The spectral pairs indicated by the dark blue markers numbered "1," "2," and "3" are spaced at 120Hz intervals, so are somehow power-supply-related; those indicated by green and red markers are spaced at 229Hz intervals, so are data-related.
No general conclusion can yet be drawn from this graph, as my experience of the Miller analyzer is still limited. However, I am currently measuring every CD-playing component available to me—from CD-ROM drives through CD players and transports to DVD players—to build up a performance database. An explanatory article will appear shortly.
While the Planet's measured performance is good, it is not really outstanding in any one area. But Roy Gandy's products have always been distinguished by sufficiently good engineering when considered across the board. The Planet is no exception. It measures as well as it needs to, as shown by the sound.—John Atkinson