21 Nov
2014
21 Nov
'14
3:42 p.m.
On Thu, Nov 20, 2014 at 11:12:09PM +0100, Ede Wolf wrote:
Thanks very much. I do know about zita-a2j any may end
up using it
for recording smtpe through onboard audio, at least as an
experiment, but for standard audio I would like to avoid resampling
whenever possible. Also, maybe pure superstition, that a2j attached
soundcard always feels as just being second best, not treated equal.
You say 'feels' and that's what it is. There is no objective reason
why resampling would reduce sound quality provided it's done as it
should be done (which is the case for ajbridge). Basically the process
is the same as the anti-aliasing filter of an AD or DA converter.
And almost all such converters include digital resampling as part of
the process.
So is there a realistic chance that I would be
presented with 52
inputs instead of 26 as of now? Without using a2j?
For synchronizing both cards I'd use external wordclock.
The next release of ajbridge will have the option to bypass
the resampling in that case. The only remaining effect would
be a somewhat higher latency for the second card (because the
periods are not synchronised).
Ciao,
--
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)