MSB Platinum Link Plus D/A processor Measurements
The MSB Platinum Link Plus locked on to S/PDIF and AES/EBU datastreams with sample rates ranging from 32kHz to 96kHz. However, at the highest sample rates, the 4x oversampling button LED changed to 2x oversampling, meaning that the DAC chips run at the same high rate when fed 44.1kHz and 88.2kHz data. The maximum output level at 1kHz was 7.046V from the balanced XLR jacks, 3.522V from the unbalanced RCAs. The latter is almost 5dB higher than the CD Standard's 2V RMS, meaning that care needs to be taken to match levels in A/B comparisons. The Platinum didn't invert absolute polarity as long as the front-panel LED wasn't lit. The source impedance was 50 ohms from the unbalanced outputs, and double that, as expected, from the XLRs. Both figures are still on the low side, however, which will minimize compatibility problems.
CD frequency response was dead flat (fig.1, top pair of traces), and while the response with de-emphasis had a very slight swayback (fig.1, lower traces), this will not have an audible consequence. Fig.1 was taken with the brickwall filter switched out of circuit. To my surprise, switching in the filter had no effect on CD response. However, it did affect the ultrasonic rolloff with 96kHz-sampled data (fig.2, top traces), extending the flat response to 42kHz. Without the filter, the response starts to roll off above 29kHz, reaching the -3dB point at 45kHz. This premature ultrasonic rolloff is associated with better-defined time-domain performance, which some commentators feel gives a more analog-like sound quality.
Fig.1 MSB Platinum Link Plus, balanced, frequency response without (top) and with de-emphasis (bottom) at -12dBFS into 100k ohms (right channel dashed, 0.5dB/vertical div.).
Fig.2 MSB Platinum Link Plus, unbalanced, frequency response at 44.1kHz (bottom at 20kHz) and 96kHz sample rates with brickwall filter switched in (top at 40kHz) and out at -12dBFS into 100k ohms (right channel dashed, 0.5dB/vertical div.).
Channel separation was better than 120dB in both directions from both sets of outputs, which is superb performance. However, traces of power-supply noise in the Platinum's outputs can be seen in fig.3. This graph shows spectral analyses of the DAC's analog output while it decodes both 16- and 24-bit data representing a dithered 1kHz tone at -90dBFS. The drop in the higher-frequency noise floor resulting from the increase in word length is almost 24dB, implying a true 20-bit dynamic range, which is extraordinary! However, the presence of tones at 60Hz (due to magnetic coupling from the AC transformer) and 120Hz (due to ripple from the full-wave-rectified power supply) interferes with the MSB's superb DAC performance at low frequencies.
Fig.3 MSB Platinum Link Plus, balanced, 1/3-octave spectrum of dithered 1kHz tone at -90dBFS, with noise and spuriae, 16-bit (top) and 24-bit data (bottom). (Right channel dashed.)
It's fair to point out that these spuriae, ranging from -117dBFS to -136dBFS, aren't going to bother anyone. But with the bar raised for DAC dynamic range by the Weiss Medea reviewed last month, that they exist at all should be pointed out.
The Platinum demonstrated superb DAC linearity, with any error (fig.4) remaining below 2dB to below -115dBFS, which is as good as the Weiss Medea DAC. As a result, the MSB's reproduction of an undithered 16-bit/1kHz tone at -90.31dBFS was essentially perfect (fig.5). Increasing the data's word length to 24 bits resulted in an excellent sinewave shape (fig.6), despite the very low signal level and the lack of linearizing dither.
Fig.4 MSB Platinum Link Plus, departure from linearity, 16-bit data (2dB/vertical div., right channel dashed).
Fig.5 MSB Platinum Link Plus, waveform of undithered 1kHz sinewave at -90.31dBFS, 16-bit data.
Fig.6 MSB Platinum Link Plus, waveform of undithered 1kHz sinewave at -90.31dBFS, 16-bit data.