My first post! 
I just read a wonderful article in (next month's?) Stereophile magazine, regarding the
Wavelength Audio USB DAC:
[Wavelength Audio designer Gordon Rankin:] "Most of these things go from the USB connection standard to an S/PDIF environment, and then to an input receiver, but I didn't want to do that. I designed my converter to go straight from the USB connection to the Philips TDA1543 chip"...
Any D/A converter and its datastream source must be synchronized with one another, and in a system of digital separates, that's usually accomplished by having the converter lock on to the word-clock signal contained in the datastream from the transport. According to Rankin, the S/PDIF environment is more prone to unwanted variations in the frequency of the incoming word-clock signal, which the converter is always trying to track. But with a USB controller generating its own word-clock signal, the frequency remains steady. "In USB," Rankin says, "the bit clock and the word clock are almost a given: There are only a few frequencies it has to deal with, and if it sees a standard PCM signal coming in ... it doesn't waver."
The interesting point here is that, properly constructed, a USB DAC hooked up to a computer should not have the timing issues that (ostensibly) motivate the expense of a high-end digital transport/outboard DAC combo.
I gather this is a controversial issue.
I've ordered my modded Teac with a USB Select jammed into it... this should be an awesome match to my Mac Mini "transport".
But do Gordon Rankin's comments apply to any USB audio interface? I gather not, as people seemed concerned about jitter in their outboard USB DACs... how much of a problem is this, really, with USB audio (and in contrast to S/PDIF coax connection)?
