LABORATORY TEST REPORT
Readers interested in a full technical appraisal of the performance of the Rotel CD11 Tribute CD Player should continue on and read the LABORATORY REPORT published on the following pages. Readers should note that the results mentioned in the report, tabulated in performance charts and/or displayed using graphs and/or photographs should be construed as applying only to the specific sample tested.
In Graph 1 Newport Test Labs has measured the harmonic distortion present in the
CD11 Tribute’s output when it’s reproducing a 1kHz signal recorded at the maximum possible level on a CD. As you can see from the graph, the third harmonic is the one that’s the highest in level, yet its level is just –95dB or (0.00177%). All the other harmonic distortion components are more than 100dB down, so each contributes less than 0.001% THD to the total.
Graph 2 shows the CD11 Tribute’s harmonic distortion levels with a 1kHz signal at –20dB, which is more typical of the maximum levels you’d find recorded on a CD, due to the need to allow headroom for peaks. You can see that the highest harmonic is now the second harmonic, at around –106dB (0.0005%). The third and fourth harmonics are both sitting around –110dB down (0.00031%) and otherwise, apart from a sixth harmonic at –118dB (0.00012%), all other harmonics are more than 130dB down (0.00003%).
Graph 3 shows performance at –60dB and although there are odd harmonic distortion components visible, they’re all more than 120dB (0.0001%) down. The ‘grass’ on the noise floor is evidence of quantisation errors, which are to be expected at this recorded level due to the absence of dither in the recorded signal.
Graph 4 shows performance with a 1kHz signal recorded at –90dB with dithering. You can see there are no distortion components visible, and there are now no quantisation errors thanks to the dithering.
SMPTE-IMD is shown in Graph 5 and you can see that the two low-frequency signals that are used (60Hz and 7kHz) are accompanied by unwanted sidebands, but all are more than 100dB down except for some sampling-related ones above 20kHz, but even the most prominent of these, up at 44.1kHz is 80dB down (0.01%).
CCIF-IMD is shown in Graph 6. The two high-frequency test signals (at 19kHz and 20kHz) have generated ten sidebands, plus there are two sampling-related artefacts at 24.1kHz at 26.1kHz. However the two highest sidebands, at 18kHz and 21kHz are at –75dB (0.0177%), the next highest are both above 25kHz and around 80dB down (0.01% THD) and all the rest are close to or more than 100dB down (0.001%).
The Rotel CD11 Tribute’s frequency response was very flat, as I’d expect, and is shown in Graph 8 with highly expanded graph scaling. You can see the CD11 Tribute’s frequency response is superbly flat out to 10kHz, after which there’s a tiny roll-off to –0.18dB at 20kHz. Excellent performance. Channel separation (not shown) was also excellent: 121dB at 16Hz, 124dB at 1kHz and 114dB at 20kHz.
Balance between the two channels was measured at a vanishingly small 0.034dB at 1kHz, again an excellent result.
Output voltage for a 0dB recorded signal was just a fraction over 2 volts, so the CD11 Tribute will have no problems driving the line input of any amplifier. Referenced to this level, Newport Test Labs measured the signal-to-noise ratios at 107dB unweighted and 119dB A-weighted, both of which are excellent results.
Linearity errors were generally low, though a bit higher than I might have expected down around the –80 to –85dB levels, but these errors seemed to correct themselves once the levels dropped to below –85dB.
All the errors tabulated in the test result chart are, however, of academic interest only, since they’re all ‘way too small to be audible, even to a highly trained ear.
Although the Burr-Brown/Texas Instruments PCM5102A DAC inside the CD11 Tribute is now almost a decade old, Rotel’s engineers have implemented it very well in this design such that overall, the Rotel CD11 Tribute CD player delivers excellent measured performance.
You can see the CD11 Tribute’s frequency response is superbly flat out to 10kHz, after which there’s a tiny roll-off to –0.18dB at 20kHz