Gramophone Dreams #69: Trance Dancing on Maxwell Street & the Rotel DT-6000 “DAC Transport” Measurements

Measurements, from March 2023 (Vol.46 No.3)

Herb Reichert reviewed the Rotel Diamond Series DT-6000 DAC Transport—fundamentally a CD player with digital inputs—in his February 2023 Gramophone Dreams column, where he described it as a “well-built, great-sounding, reasonably priced CD player.” He had so much fun using the Rotel—”rediscovering the Joy of CDs: one-box simple, stress-free, plug’n’play, and something physical to touch, scrutinize, and collect”—that he suggested I examine its measured performance.

I tested Herb’s sample of the Rotel DT-6000, serial number 272-2221004, using my Audio Precision SYS2722 system and signals I burned on a CD-R, as well as by feeding serial data to the S/PDIF and USB inputs. For the latter, I used both my MacBook Pro’s USB port, with the laptop running on battery power, and the USB port on my Roon Nucleus+ server.

The Roon 2.0 app recognized the DT-6000 as a Roon Ready device. Looking at the Rotel’s digital inputs, the optical and coaxial S/PDIF inputs accepted data sampled at all rates up to 192kHz. Apple’s AudioMIDI utility revealed that the DT-6000’s USB port accepted 16- and 24-bit integer data via USB sampled at all rates from 44.1kHz to 384kHz. Apple’s USB Prober app identified the player as “ROTEL USB Audio 2.0” from “ROTEL” and confirmed that the USB port operated in the optimal isochronous asynchronous mode.

I used the Pierre Verany Digital Test CD to check the DT-6000’s error correction as a CD player. It successfully played the tracks with gaps in the data spiral up to 1.5mm in length, but there were audible glitches when a single gap was 2mm long and with two closely spaced 1.5mm gaps. As the Compact Disc standard, the so-called Red Book, requires only that a player cope with gaps of up to 0.2mm in length, the Rotel’s error correction is superb.

The DT-6000’s single-ended output impedance was an extremely low 0.5 ohms at 1kHz and 20kHz, rising to a still very low 2.4 ohms at 20Hz. The balanced output impedance was very much higher, at close to 4k ohms at 20Hz and 1kHz, dropping to 3.65k ohms at the top of the audioband. A 1kHz signal at 0dBFS resulted in an output level of 2.155V from the single-ended outputs, which is 0.65dB higher than the CD Standard’s recommended maximum level of 2V. The balanced output level with full-scale data was 4.514V.

Fig.1 Rotel DT-6000, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).

Both outputs inverted absolute polarity, which can be seen in the Rotel’s impulse response (fig.1). This graph indicates that the player’s reconstruction filter is a short minimum-phase type, with all the Nyquist-frequency ringing following the single sample at 0dBFS.

Fig.2 Rotel DT-6000, wideband spectrum of white noise at –4dBFS (left channel red, right magenta) and 19.1kHz tone at –3dBFS (left blue, right cyan), with data sampled at 44.1kHz (20dB/vertical div.).

With white noise at –4dBFS, sampled at 44.1kHz (fig.2, red and magenta traces), the DT-6000’s response rolled off relatively slowly above 20kHz, not reaching full stop-band attenuation until the sample rate. The shape of the ultrasonic rolloff is familiar: It is typical of the filter used by MQA-capable devices for non-MQA data (footnote 1). An aliased image at 25kHz of a 19.1kHz tone at –3dBFS (blue and cyan traces) is suppressed by just 18dB, but higher-frequency images are all much lower in level. The distortion harmonics of the 19.1kHz tone all lie at or below –110dB (0.0003%).

Fig.3 Rotel DT-6000, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel green, right gray), 96kHz (left cyan, right magenta), and 192kHz (left blue, right red) (1dB/vertical div.).

Fig.4 Rotel DT-6000, spectrum of 1kHz sinewave, DC–1kHz, at 0dBFS (left channel blue, right red; linear frequency scale).

The Rotel’s frequency response with data sampled at 44.1, 96, and 192kHz is shown in fig.3. The responses all follow the same basic shape: flat to 20kHz with then a gentle rolloff disturbed by a sharp dropoff just below half of each sample rate. Channel separation (not shown) was superb, at >120dB in both directions below 1kHz and still 96dB at the top of the audioband. The low-frequency noisefloor was very clean (fig.4), with no visible power supply–related spuriae.

Fig.5 Rotel DT-6000, left channel, 1kHz output level vs 24-bit data level in dBFS (blue, 20dB/vertical div.); linearity error (red, 1dB/small vertical div.).

Fig.6 Rotel DT-6000, spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with 16-bit data (left channel cyan, right magenta) and with 24-bit data (left blue, right red) (20dB/vertical div.).

The red trace in fig.5 plots the error in the Rotel’s output level as a 24-bit, 1kHz digital tone fed to the TosLink input steps down from 0dBFS to –140dBFS. The amplitude error is less than 0.5dB until the signal lies below –120dBFS, which implies high resolution. An increase in bit depth from 16 to 24 with dithered data representing a 1kHz tone at –90dBFS (fig.6) dropped the DT-6000’s noisefloor by 18dB, which implies a resolution of 19 bits from the player’s ESS9028PRO DAC chips. This is more than sufficient to accurately decode CD data.

Fig.7 Rotel DT-6000, waveform of undithered 16-bit, 1kHz sinewave at –90.31dBFS (left channel blue, right red).

Fig.8 Rotel DT-6000, waveform of undithered 24-bit, 1kHz sinewave at –90.31dBFS (left channel blue, right red).

When I played undithered CD data representing a tone at exactly –90.31dBFS, the waveform was symmetrical, with no DC offset, and the three DC voltage levels described by the data were cleanly resolved (fig.7). Repeating the measurement with undithered 24-bit S/PDIF or USB data gave a well-formed, noise-free sinewave (fig.8).

Fig.9 Rotel DT-6000, balanced outputs, spectrum of 50Hz sinewave at 0dBFS, DC–1kHz, into 100k ohms (left channel blue, right red; linear frequency scale).

As seen in fig.2, the DT-6000 featured very low levels of harmonic distortion. The third harmonic was the highest in level from the balanced output into 100k ohms (fig.9), but this lay at just –110dB (0.0003%) and didn’t rise in level when I repeated the spectral analysis with the more challenging 600 ohm load. The third harmonic from the unbalanced output was similarly low, now joined by the second harmonic, though this lay at just –120dB (0.0001%).

Fig.10 Rotel DT-6000, balanced outputs, HF intermodulation spectrum (DC–30kHz), 19+20kHz at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

Intermodulation distortion with a full-scale mix of equal levels of 19 and 20kHz tones was extremely low in level (fig.10). The levels of the aliased images at 24.1kHz and 25.1kHz were much lower than I was expecting from fig.2.

Fig.11 Rotel DT-6000, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229.6875Hz: 16-bit CD data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

Fig.12 Rotel DT-6000, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229.6875Hz: 16-bit TosLink data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

I tested the Rotel’s rejection of word-clock jitter with the undithered Miller-Dunn J-Test signal (a high-level tone at one-quarter the sample rate over which is overlaid the least-significant bit toggled on and off at a frequency equivalent to the sample rate divided by 192). With this signal played back from CD, the Rotel reproduced the odd-order harmonics of the LSB-level, low-frequency squarewave very close to the correct levels (fig.11, sloping green line), and no other sidebands were present. However, when I repeated the spectral analysis with 16-bit S/PDIF data via TosLink (fig.12), the sidebands closest to the high-level, high-frequency tone were much higher in level. This graph was taken with optical TosLink data. The behavior with coaxial S/PDIF data was identical.

Fig.13 Rotel DT-6000, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229.6875Hz: 16-bit USB data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

The J-Test signal is not diagnostic with data where the bit clock and/or word clock are not embedded in the datastream. However, I use it to check the performance of a processor’s USB and Ethernet ports. To my surprise, the DT-6000’s USB input did poorly with this signal, whether the J-Test data were sourced from my MacBook Pro (fig.13) or from the Roon server. The presence of low-frequency audio noise broadens the base of the spectral spike representing the 11.025kHz tone, and a large number of power supply–related spuriae are present. Perhaps this is why, as HR noted in his review, the most noticeable differences between the Rotel’s CD playback and streaming files via Roon “were in energy delivery and viscosity: Music from CDs sounded denser and more fortified than music from Qobuz and Tidal. … With the DT-6000, streaming sounded less physical than phono or CD, but it was quite nice for late nights.”

With its high resolution, low noise, superbly high channel separation, and vanishingly low levels of both harmonic and intermodulation distortion, the Rotel DT-6000 offers some of the best measured performance I have encountered from a CD player. However, because of its high levels of jitter, as a streaming DAC, its performance is not in the same class.—John Atkinson

Footnote 1: See, for example, fig.2 here.

Page 1
Gramophone Dreams #69: Trance Dancing on Maxwell Street & the Rotel DT-6000 "DAC Transport" Page 2
Gramophone Dreams #69: Trance Dancing on Maxwell Street & the Rotel DT-6000 "DAC Transport" Measurements

Click Here: classic gaa shirts