In Reply to: RE: Merging HAPI as digital transport posted by Sprezza Tura on October 29, 2014 at 11:02:00:
A "computer" of some sort is required to implement the required protocols over Ethernet. (I put quotes in so as to try to avoid revisiting an earlier debate over terminology.) This "computer" will need to connect via some means to the actual DAC circuitry (e.g. digital logic possibly including DSP, clock, digital to analog switches, IV and analog output buffer). The "computer" will likely be a system on a chip (SOC) and it will connect to the DAC via some standard method, which may be I2S or may be USB. The firmware for this computer will probably be burned into the SOC's flash memory, and possibly subject to being updated.
So introducing Ethernet isn't really changing anything when it comes to noise and jitter, all the same "computer" issues will apply between the SOC and the DAC, as Thorsten points out. For studio use there are obvious benefits in terms of flexibility, but these probably won't apply for many audiophiles who have a single high-end system. How good the sound quality is ultimately will depend on the quality of implementations, e.g. what's the programming in the SOC, how are the power and ground handled, etc... In particular, based on past experience, I would be surprised if the clock implementations didn't turn out to be worse than with previous methods, at least in initial product releases.
tl;dr
The devil is in the details.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- RE: Merging HAPI as digital transport - Tony Lauck 10/30/1408:19:47 10/30/14 (1)
- RE: Merging HAPI as digital transport - Sprezza Tura 08:31:27 10/30/14 (0)