This review page is supported in part by the sponsors whose ad banners are displayed below's office automatically generated and emailed out a sales invoice for the review loaners. With $124.26 for globally expedited UPS plus a super stiff $250 termination labor—whoa!—it totaled $1.235.15. The Pro's outputs still alien to consumer hifi come at a price. My invoice included two cable sets of course to run balanced or single-ended. Civilians would only need one cable set. Using your own wires eliminates such cable fees. But, a pair of Cardas or other Tiny-Q to XLR or RCA adaptors seems to be an unavoidable surcharge even then. Reader Roscoe 'Trey' Nicholson sourced a $40 pair of 4' cables from kV Connection. "Very quick turnaround on cable pricing and construction. The Mogami W2330 used has a quite small 28-gauge diameter. I had requested Mogami 2549 but was told its was too large for the mini-XLR connectors they had. Even with the thin Mogami, there was clear improvement over what I heard in the Streamer II even though the II ran a superior Tara Labs RSC Reference Gen 2 cable."

Inventory # Item description Unit retail Number of items Line total
9003379 Furutech FP-701 M(G) high-performance male XLR $12.55 2
9003383 Furutech FP-120 F(R) filament-central PIN RCA plug $35.00 2
9003498 Furutech U-P2.1 balanced interconnect cable (50m/R) by the foot $42.11 3
9006062 Furutech cable termination fee $125.00 2
9009158 HRT MusicStreamer Pro $499.95 1
9009204 Switchcraft Tiny-Q Mini XLR female plug $13.00 4
      Sub total
      UPS worldwide

Running my Weiss DAC2 and the loaner Streamer off the iMac—one via Firewire, the other via USB—the former was unceremoniously bumped from the selectable audio options in the OSX sound devices window. Kevin Halverson guessed that because the Weiss uses custom drivers from its installation disc but his Pro native drivers, the connected Streamer made my OS default to the factory (native driver) configuration. Disconnecting the Streamer and rebooting the Weiss had the latter shake hands again with its own driver software. While this made comparisons less convenient than imagined, they were manageable once I understood proper procedure. In real life, it's unlikely that home audio customers would have multiple DACs with different interfaces hanging off the same host computer.

Kevin's brief primer on USB power: "Some portables limit their bus power in order to conserve battery life. A proper USB host will supply 500mA per downstream device. A passive hub will distribute this power to all downstream ports, often reducing power to 100mA per. You may have noticed that some bus-powered USB peripherals (particularly HDD and optical drives) supply a Y cable. The goal is to allow the 500mA limit to be exceeded by combining two ports just for power. In the case of the Music Streamer Pro, it requires up to 480mA depending upon the sample rate. Most hosts can supply this without any problem unless a passive hub is involved. If the host is configured to supply less than a normal full load, then a self-powered USB hub is necessary to overcome the limit. This applies equally to all current Streamers as they all require more than 100mA (Streamer II = 200mA, Streamer II+ = 350mA).

"There is a second mechanism which is completely unrelated to power when a hub can provide a performance advantage. The mechanism for this improvement is complex but a simplified description is that the addition of a hub can offload one task from the host computer and depending upon the load—computational and otherwise—provides an advantage. Again, this isn't related to power at all and only comes into play in a small percentage of hosts. Lastly, with an asynchronous device nearly all mechanisms the cable can contribute to sonics are eliminated. The only one remaining (presuming a properly executed cable) is shielding. Here the quality of the shield and the resultant impact on other components in the system are a complex variable, susceptibility of the other devices being the major contributor. There are a few things that fall into the general interest category worthy of consideration about USB—or for that matter, any data—cables. 1/ No cable improves with length. In the case of USB, there are fairly hard limits of 5 meters per segment. This limit can be broken up to 5 times for a total length of 30 meters if active data repeating is utilized at each 5 meter point. 2/ Try not to run data and analogue audio cables in parallel. This unnecessarily increases the coupling between them and can diminish performance. 3/ If data and analogue cables must come near one another, try to have them cross at right angles to minimize the coupling mechanism."

On parts sleuthing across the two MS Pro boards—the mother board higher up and the piggy-backed daughter board right above—you would identify the TAS 1020N USB transceiver, a PCM 1794 DAC and chips including the 5534 (x 4), 4121 (x 2), 78L05, 24C64WI (socketed), 86DN79T, MIC 4468ZWM and 2C32A. There were others whose writings I couldn't decipher. Their various functions in the circuit remain HRT's secret. In matters of public recognition however, the Streamers are far from a secret. This Swedish hair salon runs an upscale stereo system with six wall-mounted Gallo Stradas augmented by one Gallo TR3 sub. The front end is a playlisted PC with HRT Streamer and Entreq USB cable. On Friday afternoons, a disc jockey heats up the joint while customers relax with a glass of bubbly. Now that's cross fertilization in action - and precisely the concept brief and rationale of the Streamers' stripped-down cosmetics, focused functionality and aggressive pricing. Make friends outside entrenched audiophilia.