cliveb;656138 Wrote: 
> Yes, that is precisely the situation I'm referring to. If you want to
> deliver your final product at 44.1kHz, at some point you have to
> eliminate everything above 22.05 to avoid aliasing. Doesn't matter if
> you start by recording at a high sample rate to avoid the need for a
> brick-wall: as soon as you downsample you have to filter.
I think we are in agreement- it's just the expression "brick wall"- I
understadn this to mean a filter with effectively no transitional band
ie one which provides infinite attentuation at the transition between
the pass band and the stop band. In practice this may mean something
with really steep attenuation at the transition. This is really
difficult to do in analog but is trivial in the digital domain because
all you have to do is to calculate the values that the samples would
have had *if* a perfect analog filter existed.

I suppose you could say that the combined effect of the digital
filtering and the analog filter is "brick wall", but the brick wall bit
is done digitally. This leaves the analog filter only having to do the
same job it would have done if the final output was going to be at the
oversampled rate. 

I guess this comes back to your point that even a digital brick-wall
filter has some (perceptible) phase effect. if that's right then I
agree that there would be an advantage in having a final product with
fs > 44.1kHz (but >96kHz?) 
Wombat has answered that point, but i have to say that i have not yet
quite got to grips with that bit. Certainly i am aware that recent
hi-end products claim to have produced filters which solve a problem
called pre-ringing. I am at the moment agnostic about whether this is a
real solution to a real problem, but my suspicion is that it is yet
another example of a solution for the imaginary shortcomings of
"digital"


-- 
adamdea
------------------------------------------------------------------------
adamdea's Profile: http://forums.slimdevices.com/member.php?userid=37603
View this thread: http://forums.slimdevices.com/showthread.php?t=89733

_______________________________________________
audiophiles mailing list
audiophiles@lists.slimdevices.com
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to