Halide Design S/PDIF Bridge USB-S/PDIF converter Measurements
Sidebar 3: Measurements
I examined the measured behavior of the BNC version of the Halide USB converter using the Audio Precision SYS2722 system (see www.ap.com and "As We See It" in the January 2008 issue), as well as the Miller Audio Research Jitter Analyzer. I played test tones at various sample rates and bit depths from BIAS Peak Pro 6 on an Intel MacBook running OS10.6.4, with some tests repeated on a dual-core PC running Windows 7 and Adobe Audition 3.0.
The Halide operated at whatever sample rate the audio file had been recorded at, provided the sample rate had been set with Audio MIDI Setup on the Mac or Audio DevicesProperties on the PC. Very importantly, the Halide Bridge did operate at the 88.2kHz sample rate. I checked that the Halide was bit-transparent by feeding its S/PDIF output to an RME soundcard fitted to a second PC, and recorded the data using Adobe Audition. When I played that recording simultaneously with an inverted, bit-synchronized version of the original audio file, there was a perfect null with both 16- and 24-bit data, proving that the bits output from the converter via S/PDIF were the same bits sourced from the host computer via USB. I used RME's DIGICheck utility to examine how many bits were active in the Halide Bridge's S/PDIF outputthe number of active bits followed how many had been set in the Mac's Audio MIDI Setup utility or the PC's Audio DevicesProperties dialog: 16 or 24, as appropriate.
The USB Prober program revealed that the Halide Bridge operated in asynchronous isochronous mode, as specified. The "eye pattern" of the S/PDIF data waveform was wide open and free from timing uncertainty at its start and end (fig.1), and the Audio Precision System SYS2722 calculated the jitter in the S/PDIF datastream to be a very low 345 picoseconds peak. For reference, the other three USBS/PDIF converters I have testedthe Bel Canto USB Link 24/96, Lindemann USB-DDC 24/96, and Stello U2respectively measured 2.91 nanoseconds (2910ps), 444ps, and 395ps, all with a 50Hz100kHz measurement bandwidth.
Fig.1 Halide Design S/PDIF Bridge, eye pattern of S/PDIF data output carrying 16-bit J-Test signal (±500mV vertical scale, 175ns horizontal scale).
I looked at the effects of datastream jitter in the reconstructed analog signal with a 1995-vintage Assemblage DAC-1 D/A processor, sourcing 16-bit J-Test audio data from the MacBook's USB output. I chose the Assemblage because it appears to have the worst rejection of incoming datastream jitter of the DACs I had to hand.
Fed S/PDIF data via TosLink from the RME soundcard in one of my test-lab PCs, the spectrum of the Assemblage's analog output suffered from very high levels of data-related sidebands, as well as significant broadening of the central peak that represents a high-level tone at exactly one-quarter the sample rate (fig.2). The Miller Analyzer calculated the jitter level to be an enormous 11.4ns peakpeak. By contrast, feeding the Assemblage data from the Halide S/PDIF Bridge reduced the jitter by a factor of 10, to 1.1ns pp. While there is still some accentuation of the lower-frequency data-related sidebands, the spectrum was significantly cleaner, with now a well-defined central spike (fig.3).
Fig.2 Assemblage DAC-1, high-resolution jitter spectrum of analog output signal, 11.025kHz at 6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz, S/PDIF data from RME soundcard via 15' TosLink. Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz (left channel blue, right red).
Fig.3 Assemblage DAC-1, high-resolution jitter spectrum of analog output signal, 11.025kHz at 6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz, S/PDIF data from Halide Design S/PDIF Bridge. Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz (left channel blue, right red).
Replacing the Assemblage with the Musical Fidelity X-24K DAC gave a similar reduction in jitter when the TosLink connection was replaced by the Halide Bridge, with the latter offering just 185ps pp in the reconstructed analog waveform, and a much cleaner spectrum with all but the lowest-frequency data-related sidebands at the residual level (fig.4). For reference, the Bel Canto USB Link gave 4.57ns pp with the X-24K, the Stello U2 455ps, and the Lindemann USB-DDC 270ps.
Fig.4 Musical Fidelity X-24K, high-resolution jitter spectrum of analog output signal, 11.025kHz at 6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz, S/PDIF data from Halide Design S/PDIF Bridge. Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz (left channel blue, right red).
Tested with the Esoteric D-07, with which I thought the Halide Bridge worked well, the TosLink output of my MacBook gave 1049ps pp; the MacBook via the Stello U2 actually increased this slightly, to 1090ps, while the Halide Bridge reduced the measured jitter to 780ps. This indicates that the Halide's Streamlength code works as advertised.John Atkinson