Musical Fidelity M1 DAC John Atkinson January 2013
John Atkinson reviewed the Musical Fidelity M1DAC in January 2013 (Vol.36 No.1):
Sam Tellig enthusiastically reviewed this inexpensive solid-state D/A processor in March 2011. "What I mainly heard from the M1DAC was nothing: an absence of artifacts, if you want to get fancy," he wrote. "There was no fudging of detail, no smearing of transients. Purity of tone was exceptional."
Some months after that review was published, the M1DAC was updated. Whereas the original version's USB input was limited to 48kHz sample rates and below and 16-bit data, the new version uses the popular Texas Instruments TAS1020B USB receiver, operating in the optimal asynchronous mode, to handle 24-bit data with sample rates up to 96kHz. A few other things also changed, mainly concerning the power supply, but otherwise the circuit remains the same, with a Burr-Brown DSD1796 DAC chip converting the bits and sending the analog signal to an output stage based on a JRC 5532 dual op-amp chip. The price has increased, by $50, to a still-reasonable $749. You can find Michael Lavorgna's thoughts on the sound quality of the new version of the M1DAC here.
The serial number of the original sample that I measured was MDY0085. The new sample I examined for this Follow-Up was MDY0648. Apple's USB Prober utility reported the M1DAC's manufacturer as "Musical Fidelity," and gave the product string as "Musical Fidelity M1DAC 24/96," operating in isochronous asynchronous mode as claimed, and capable of handling 24-bit data with sample rates of 32, 44.1, 48, 88.2, and 96kHz.
I examined the Musical Fidelity's electrical performance with Stereophile's loan sample of the top-of-the-line Audio Precision SYS2722 system (see www.ap.com and the January 2008 "As We See It"; for some tests, I also used my vintage Audio Precision System One Dual Domain. Most measurements of the new samplemaximum output level, output impedance, non-inverting polarity, channel separation, distortion spectrumwere the same as or very similar to those of the old sample. The frequency response, for example, was the same (fig.1), with 192kHz data (blue and red traces) still not giving a greater bandwidth than 96kHz data (cyan and magenta).
But one aspect was significantly improved. Although it had impressive resolution with 24-bit data, the original sample's output was marred by low-level power-supply spuriae (fig.2). The new sample, measured in an identical manner with S/PDIF data, gave a spectrum free from such spuriae (fig.3). The increase in bit depth from 16 to 24 dropped the noise floor by 22dB, even via USB (fig.4), implying a resolution close to 20 bits, which is excellent.
Fed a 24-bit version of the Miller-Dunn J-Test data via TosLink, the M1DAC's output spectrum was free from jitter-related artifacts (fig.5). A very similar result was obtained via USB (fig.6), confirming the new USB receiver chip's superb jitter rejection.
The new version of the M1DAC resolves the slight measured problems of the original and can be confidently recommended.John Atkinson