Page 2: Ferrofish Pulse 16 DX

Technology

Not only does the Pulse 16 look very similar to the A32, it also shares much of the same internal technology. The converter chips are the same as those used in the A32, with Cirrus Logic CS5368 A-Ds and the complementary CS4385 D-As — both being eight-channel devices, and both specified with a 114dB dynamic range. Also, the supporting analogue circuitry is very similar, based upon Texas Instruments RC4580 dual op-amps. The Dante interface in the DX model uses Audinate's familiar Brooklyn-II module (as do all of Ferrofish's Dante products), ensuring reliable compatibility with other manufacturers' Dante equipment. A Sharc micro-controller takes care of all the signal routing, levels, and configuration settings.

I obtained a measurement of 112.2dB (A-wtd) for the A-D, which is roughly 6dB better than the figure I achieved with the A32 and 3dB better than the old A16 Ultra.

Running my standard AES17 dynamic range test using an Audio Precision test set, I obtained a measurement of 112.2dB (A-weighted) for the A-D, which is roughly 6dB better than the figure I achieved with the A32, and 3dB better than the old A16 Ultra. For the D-A the AES17 figure came out as 111.5dB (A), which is about the same as both the A32 and A16 Ultra results. Although the state of the art achieves dynamic range figures in the low 120s, these are quite reasonable and acceptable results for a device of this type and cost, and very unlikely to have any negative practical impact in typical applications.

All of my other bench tests came out very well, with IMD measuring 0.004 percent (D-A) and 0.003 percent (A-D), THD+N measured 0.003 percent (both D-A and A-D), crosstalk at both 1 and 10 kHz was -107dB (D-A) and -109dB (A-D), and the frequency response is ruler flat from below 5Hz to 23kHz when operating with a 48kHz sample rate. Attenuation at the Nyquist frequency measured around -15dB, which, although not strictly adhering to the Nyquist requirements, is better than many converters which provide only -6dB! The practical impact of this filter response is a modest potential for aliasing distortions if recording very loud signals with very strong harmonics... which isn't that common a requirement.

Impressions

The Pulse 16 is an elegant update on the previous A16 Ultra converter, with slightly improved technical performance. The model range provide useful variants for those that want MADI and/or Dante interfacing, while keeping the price as low as possible for those that just want a simple cost-effective ADAT expander for line-level analogue equipment. Its physical styling also sits well with the A32, giving a more homogenous look to the entire Ferrofish product portfolio.

Adding Dante capability is a smart move since this is rapidly becoming the de facto audio-over-IP format in studio, broadcast, and live-sound applications. However, until Dante completely replaces MADI as the preferred multichannel audio distribution network, being able to exchange audio between the two formats via the Pulse 16 DX is a very useful facility.

The Pulse 16/MX/DX converters should do extremely well as they are not only very well-engineered products providing highly cost-effective and capable multichannel ADAT/analogue converters, but they also offer access to the MADI and Dante formats at a price which is significantly below that of the competition. And while the converter performance may be a few decibels below that of the leading high-end products, I doubt anyone would notice or care in most real-world applications.