USB yet again


For a few years I have had an Intona Isolator with Oyaide Continental 5S upstream and Intona Reference downstream connecting Streamer and DAC. Given the Strong benefit of filters on the upstream Ethernet connection I added a LHY Audio USB 3.0 purifier with a Grey Knights power cable. The tightening of the transfer and resultant SQ was remarkable despite having used superior cables before.

While USB remains a compromised transfer format, asynchronous USB is the only protocol synchronising the server’s and dac’s clocks unless both have master clock connections. AES/EBU may have better noise rejection but has imbedded clock signal,SPDif is outdated as well as speed constrained and I2S not standardised. Hopefully the industry comes up with a better solution. It is interesting that there seems to emerge a trend to combine server and dac: one wonders why?

antigrunge2

Showing 6 responses by antigrunge2

@lalitk 
 

agree with your assessment. It seems though pretty damning that both on Ethernet and USB even on high end equipment you have to engage in extensive optimisation efforts before you get a decent result. As I said the single box solution can be seen as a capitulation.

@itsjustme

technically speaking it slaves the server’s clock, i.e. data is requested by the dac rather than the server sending it on its own clock.

I stated that SPDif is speed constrained, not AES/EBU.

@sns 

my point, precisely. I also steuggle with dependence on cables and cleaned up power

@designsfx 

I am not aware of any SPDif connections faster than 192KHz, conversely double AES/EBU can run at 384MHz

@vthokie83 

A DDC reclocks the USB connection. Therein may lie the principle benefit to SQ, probably not in using another port on the DAC.