GamuT CD 1 CD player Measurements
The unbalanced RCA output jacks were loose on the review sample, but still made good electrical contact. The GamuT player offered a maximum output of just over 4V RMS from both balanced and unbalanced jacks, 6dB higher than the CD standard's 2V. (Both balanced phases were active.) Neither output inverted absolute polarity. The source impedance was a low 56 ohms across the audioband from the unbalanced RCAs, 150 ohms from the balanced XLRs. Error correction was excellent, the CD 1 coping with gaps in the data spiral of up to 1mm in length without stuttering.
The GamuT's frequency response was different from its two pairs of outputs, the unbalanced signal rolling off by 1.1dB at 20kHz compared with the balanced's 0.25dB (fig.1). This might correlate with BD's finding the balance a little on the warm side. The response with pre-emphasized data (offset by -1dB in fig.1 for clarity) was shelved down by 0.1dB above 2kHz, which will be just audible as a slight dulling of the sound. Channel separation (not shown) was superb at better than 110dB in the midrange, but a little less good at high frequencies from left to right than from right to left.
Fig.1 GamuT CD 1, frequency response at -12dBFS into 100k ohms (from top to bottom): balanced outputs, unbalanced outputs, balanced outputs with de-emphasis (right channel dashed, 0.5dB/vertical div.).
Fig.2 shows a 1/3-octave spectral analysis of the CD 1's output while it decoded dithered data representing a 1kHz tone at -90dBFS. No power-supply or harmonic spuriae can be seen, and the noise floor basically consists of the dither used to linearize the recorded signal. This graph was taken from the unbalanced outputs. Interestingly, repeating the spectral analysis from the balanced jacks revealed the presence of some 60Hz hum, though at almost 120dB below the maximum level (not shown), it will not have audible consequences. A wider-band spectrum of the GamuT's noise floor while it played back data representing a -1LSB DC offset (fig.3) was very clean, though a very small amount of energy centered on 44.1kHz can be seen in the left-channel output.
Fig.2 GamuT CD 1, unbalanced, 1/3-octave spectrum of dithered 1kHz tone at -90dBFS, with noise and spuriae (16-bit data, right channel dashed).
Fig.3 GamuT CD 1, unbalanced, 1/3-octave spectrum of -1LSB, with noise and spuriae (16-bit data, right channel dashed).
The GamuT's Crystal DAC has low linearity error down to below -100dBFS, as can be seen from fig.4, though a slightly higher level of high-frequency noise than is usual these days can be seen to affect the graph below -110dBFS. This slightly obscures the tri-level shape of an undithered 1kHz tone at -90.31dBFS (fig.5) compared with the best-performing digital gear.
Fig.4 GamuT CD 1, left-channel departure from linearity, 16-bit data (2dB/vertical div.).
Fig.5 GamuT CD 1, waveform of undithered 1kHz sinewave at -90.31dBFS, 16-bit data.
The CD 1's analog output stage appears to be based on an Analog Devices AD712 dual BiFET op-amp chip followed by a Burr-Brown OPA2134 low-distortion, dual-op-amp chip, the latter optimized for audio work. The analog circuitry was superbly linear even into low impedances. This can be seen in fig.6, which reveals that, with the CD 1 driving a full-scale 1kHz tone into 8k ohms, all the distortion harmonics are below -100dB (0.001%), with the third harmonic being the highest in level at -101dB, and the right channel (red trace) being very slightly less good than the left. Intermodulation distortion (fig.7) was also superbly low in level, with the 1kHz difference component lying at -99dB!
Fig.6 GamuT CD 1, unbalanced, spectrum of 1kHz sinewave, DC-10kHz, at 0dBFS into 8k ohms (linear frequency scale).
Fig.7 GamuT CD 1, unbalanced, HF intermodulation spectrum, DC-25kHz, 19+20kHz at 0dBFS into 8k ohms (linear frequency scale).
The Miller Audio Research Jitter analyzer revealed that the CD 1 produced low levels of word-clock jitter: 253 picoseconds peak-peak. The spectrum of the noise floor around a high-level tone at exactly one quarter the sample rate is shown in fig.8. (The data are modulated at 229Hz with an amplitude of 0.5LSB.) Data-related sidebands are indicated with red numeric markers and are all low in level, even at the fundamental frequency of ±229Hz (red "3"). However, the strongest sidebands are at the power-supply-related frequency of ±120Hz (blue "2") and at the unusual frequency of ±1770Hz (purple "16"). The central tone also shows the spectral spreading that appears to be typical of sigma-delta DAC topologies. I do wonder if this slight spreading, in combination with the low frequency, correlates with BD's feeling that the GamuT CD 1 had a rather warm presentation.
Fig.8 GamuT CD 1, unbalanced, high-resolution jitter spectrum of analog output signal (11.025kHz at -6dBFS sampled at 44.1kHz with LSB toggled at 229Hz). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.
All in all, this is excellent measured performance.—John Atkinson