Antelope Audio Zodiac Platinum D/A processor–headphone amplifier & Voltikus power supply Measurements

Sidebar 3: Measurements

I examined the Antelope Audio Zodiac Platinum's electrical performance with my Audio Precision SYS2722 system (see www.ap.com and the January 2008 "As We See It"). Unless stated otherwise, all measurements were performed from the Zodiac's balanced outputs. I don't yet have files of DSD-encoded test signals, so the Zodiac's measured performance was assessed with PCM-encoded files, with sample rates extending from 32kHz up to 384kHz. Most of the testing was performed from the Antelope's balanced outputs—selected by holding down the output button for more than 3 seconds, which illuminates the triangle of white LEDs—with upsampling turned off.

The Antelope's performance via its USB port was tested using my 2012-vintage Apple MacBook Pro. Apple's USB Prober utility identified the Zodiac Platinum as having the product string "Zodiac Platinum" and the manufacturer string "Antelope Audio," and confirmed that the processor's USB input operated in isochronous asynchronous protocol. The AudioMIDI utility listed all sample rates from 32 to 384kHz as being possible via USB, with 24-bit word lengths. The TosLink inputs locked to datastreams with sample rates up to 96kHz; the coaxial and AES/EBU inputs operated successfully with datastreams up to 192kHz.

The Zodiac's default settings, confirmed with the Mac OSX Control Panel app, were 20dBU for the balanced outputs, 6dBV for the unbalanced outputs, and 0dB attenuation for the headphone outputs. The maximum output level at 1kHz was 7.79V from the balanced and headphone outputs, and 1.961V for the unbalanced outputs. All the outputs preserved absolute polarity (ie, were non-inverting). The output impedances didn't vary from 20Hz to 20kHz, and were all low: 112 ohms balanced, 56 ohms unbalanced, 121 ohms with the headphone output set to "120," and 3 ohms with it set to "0." (All the measured impedances include the series impedance of 6' of cable.)

Fig.1 shows the Zodiac's impulse response with 44.1kHz data; it reveals the reconstruction filter to be a time-symmetrical, finite-impulse-response type. Interestingly, unlike with some other upsampling processors I've tested, activating the Zodiac's upsampling to 352.8kHz gave a much shorter impulse response (fig.2). With 44.1kHz-sampled white noise (fig.3, magenta and red traces), this filter rolled off the output rapidly above 20kHz, with a notch at the Nyquist Frequency (half the sample rate, indicated in the graph with a green vertical line). With a full-scale 19.1kHz tone (cyan and blue traces) there was no trace of the ultrasonic image at 25kHz, and the distortion harmonics all lay close to or below –100dB. Examined with the volume pot set to "0," the Zodiac's frequency response with sample rates ranging from 44.1 to 384kHz is shown in fig.4. The response follows the same pattern with all four sample rates: flat to 20kHz, with a gentle rolloff above that frequency, on which is overlaid a sharp rolloff just below each Nyquist Frequency. Note the superb channel matching in this graph.

914AntZodfig01.jpg

Fig.1 Antelope Zodiac Platinum, impulse response at 44.1kHz (4ms time window).

914AntZodfig02.jpg

Fig.2 Antelope Zodiac Platinum, impulse response at 44.1kHz upsampled to 384kHz (4ms time window).

914AntZodfig03.jpg

Fig.3 Antelope Zodiac Platinum, wideband spectrum of white noise at –4dBFS (left channel magenta, right red) and 19.1kHz tone at 0dBFS (left cyan, right blue), with data sampled at 44.1kHz (20dB/vertical div.).

914AntZodfig04.jpg

Fig.4 Antelope Zodiac Platinum, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel yellow, right gray), 96kHz (left blue, right red), 192kHz (left cyan, right magenta), 384kHz (left blue, right red) (2dB/vertical div.).

Channel separation (not shown) was excellent if somewhat asymmetrical, with the R–L crosstalk rising from –120dB at 500Hz to –87dB at 20kHz, while in the other direction it remained below –110dB at 20kHz. Examining the low-frequency noise floor while the Zodiac reconstructed a full-scale 1kHz tone (fig.5) showed that there were no power-supply-related spuriae above –145dB!

914AntZodfig05.jpg

Fig.5 Antelope Zodiac Platinum, spectrum of 1kHz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

With dithered 16-bit data describing a 1kHz tone at –90dBFS (fig.6, cyan and magenta traces), the traces peak at exactly –90dB, implying superbly low linearity error and the noise floor is actually that of the dither used to encode the data. With dithered 24-bit data (blue and red traces), the noise floor drops by 27dB, suggesting that the Zodiac Platinum offers resolution of between 20 and 21 bits, which is both the state of the art and easily enough to fully resolve a tone at –120dBFS (fig.7). This graph was taken with AES/EBU data; repeating the analysis with USB data confirmed that the Zodiac's USB input correctly handles 24-bit data. With its very low noise and superb resolution, the Zodiac's reproduction of an undithered tone at exactly –90.31dB was essentially perfect (fig.8). Not only is the waveform superbly symmetrical, with the three DC voltages described by the data clearly evident; the Gibbs Phenomenon "ringing" at the waveform transitions is very well defined. With undithered 24-bit data (fig.9), the result is a superbly clean sinewave, despite the very low signal level and the lack of dither.

914AntZodfig06.jpg

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

914AntZodfig07.jpg

Fig.7 Antelope Zodiac Platinum, spectrum with noise and spuriae of dithered 1kHz tone at –120dBFS with 24-bit data (left blue, right red) (20dB/vertical div.).

914AntZodfig08.jpg

Fig.8 Antelope Zodiac Platinum, waveform of undithered 1kHz sinewave at –90.31dBFS, 16-bit data (left channel blue, right red).

914AntZodfig09.jpg

Fig.9 Antelope Zodiac Platinum, waveform of undithered 1kHz sinewave at –90.31dBFS, 24-bit data (left channel blue, right red).

With full-scale signals, the Zodiac Platinum was not quite as linear as the best-measuring DACs. Fig.10 shows the spectrum of its output while driving a 50Hz tone at 0dBFS into 100k ohms. Some low-level, high-order harmonics are visible, and the second harmonic is the highest in level in the left channel (blue trace), the third in the right channel (red). However, at close to –100dB (0.001%), these are still very low in absolute terms and, commendably, they didn't increase their level when I reduced the load to the punishing 600 ohms. Lowering the signal level by 10dB eliminated most of the higher-order harmonics and dropped the second and third harmonics by 6dB (fig.11). Intermodulation distortion was also very low (fig.12).

914AntZodfig10.jpg

Fig.10 Antelope Zodiac Platinum, spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

914AntZodfig11.jpg

Fig.11 Antelope Zodiac Platinum, spectrum of 50Hz sinewave, DC–1kHz, at –10dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

914AntZodfig12.jpg

Fig.12 Antelope Zodiac Platinum, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

I tested the Zodiac Platinum's rejection of word-clock jitter with the TosLink, AES/EBU, and USB inputs. Despite the Miller-Dunn J-Test signal's not being diagnostic for USB data, where the clock is not embedded in the datastream, the results were identical for all three inputs. Fig.13 shows a narrowband spectrum of the Antelope's output while I decoded 16-bit J-Test data via TosLink. The odd-order harmonics of the low-frequency, LSB-level squarewave are mainly close to their correct levels (green line), but there are some differences in the levels of the harmonics closest to the spectral spike, which represents the high-level tone at 11.025kHz. There is also a spurious tone present just below 12kHz, at –124dB.

914AntZodfig13.jpg

Fig.13 Antelope Zodiac Platinum, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 16-bit data via TosLink from AP SYS2722 (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

Repeating the spectral analysis with 24-bit J-Test data (fig.14) eliminates all the odd-order harmonics, as expected, but as well as the spurious tone, a pair of sidebands is now visible at ±229.6875Hz. It is the interaction between these tones and the odd-order harmonics that changes the levels in fig.10. As these sidebands are also present with 24-bit USB data, they are unlikely to be due to jitter. Perhaps they arise from a small amount of Meitner and Gendron's Logic-Induced Modulation. Figs. 10 and 11 were taken without upsampling; switching in upsampling to 352.8kHz didn't change the major aspects of these graphs, though it did give rise to some low-level sidebands sticking up about the 24-bit noise floor (fig.15).

914AntZodfig14.jpg

Fig.14 Antelope Zodiac Platinum, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 24-bit data via TosLink from AP SYS2722 (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

914AntZodfig15.jpg

Fig.15 Antelope Zodiac Platinum, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 24-bit data via TosLink from AP SYS2722 upsampled to 384kHz (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

Finally, looking at the Zodiac Platinum's analog inputs, the unbalanced input was non-inverting and offered unity gain. The input impedance was low, at 3k ohms across the audioband, and the response rolled off above that band, reaching –2dB at 20kHz and –6dB at 50kHz (fig.16). Distortion for analog input signals was very low, consisting primarily of the third harmonic at –103dB (not shown).

914AntZodfig16.jpg

Fig.16 Antelope Zodiac Platinum, unbalanced analog input, frequency response at 1V into 100k ohms with volume control at "0" (left channel blue, right red) (1dB/vertical div.).

Overall, the very good measurements of Antelope Audio's Zodiac Platinum indicate excellent digital and analog engineering.—John Atkinson

COMPANY INFO
Antelope Audio
San Francisco, CA
(415) 869-9661
ARTICLE CONTENTS

COMMENTS
tonykaz's picture

Hello Sir ,

Our Tyll is revealing the newish strategy at your offices , I'm happy to learn of your position as the "Man" covering DACs .

I've had a look-see at your MSB writings as well as a few other offerings , certainly I'll continue to follow along .

I'm an old time Audio person , going way back to the early Tube days . I've imported some of the British products ( back in the 1980s ) , I've retailed and manufactured .

Now-a-days I'm a Schiit/Sennheiser music lover , I have no business connection to Home or Professional music reproduction .

I've come to discover DACs to be a sort of Phono Cartridge type of transducer , the MSB stuff seems the best of the bunch .
Not being in the business , I don't have any consideration other than the performance of the device in question ( a happy place ) .

I think that I've discovered the Dopamine connection to music and it's reproduction , this phenonomen occurs when any music played ( poorly or well ) excites the brain into a state of pleasantness , I suppose it's why we all love music .

I think you lads have a million things to consider when reviewing products , I don't expect reviewers to get down to the tiny details of how a femto clock can keep music in focus or things of that nature but I will expect honesty and integrity , which I suspect your JA will insist upon ,

Anyway , thank you for pitching in and trying to be helpful in describing the advancements in this area .

Tony Kaz in Michigan

ps . your local Dr.Mike Moffet might be helpful in digital education , of course the MSB lads are close by in Watsonville so you do have potential access to the pointy headed intellectuals .

X