-=> Quoting Gordon Gilbert to T Owen <=-
 GG> Ok, so you agree 16-bit is fine and we don't really need 20 or
 GG> 24 bit CDs in the home.  We're halfway there....
I wouldn't know about in the home, since I don't have a CD player at
home. There are probably folks that listen to enough music in the home,
with quiet enough equipment that they might buy a 24 bit CD player, but
it wouldn't be me.
 TO> use the maximum resoloution of the medium. Dynamic range is not
 TO> the problem with CD; it is the lack of resoloution.
 GG> What do you mean by resolution exactly?  The CD handles the
 1Hz -> 21.8kHz region just as well as it could if the sampling rate
 GG> was 100,000 times the current rate.  You would only gain frequency
 GG> response.
Frequency response is precisely what I am talking about; the wider, the
better.
       
 GG> Show me a scientific study that proves this is true.  Pointing
 GG> at a square wave at 20kHz does not constitute adequate proof.
If there was scientific proof, I would gladly post it. Since it is a
matter of subjective experience, I cannot show proof that we *need*
wider than 1-21.8KHz, but I have found that music sounds more "open"
and "transparent", more natural, when response is extended well beyond
20kHz.
 GG> At what point should we call it
 GG> adequate?  You suggested 10 times in a previous message.  That would
 GG> mean response to over 200kHz.  Do you seriously think sounds in that
 GG> range affect our hearing?
No, I don't think that we need to go that far with *reproduction* of
sound, but the bandwidth is important, at least in the analog recording
and repro chain. Digital that is *capable* of 200kHz would be very
accurate in the frequency ranges that we can hear, and those slightly
above.
 GG> Do you seriously think we have microphones
 GG> that will accurately record that range?
No, nor did I claim anything of the sort.
 GG> There are *very* few that
 GG> will record flat to 20kHz, let alone 48kHz or beyond.
"Flat", of course, is a relative term; some folks think that +/- 3db
is flat. 
 GG> Frankly, short of scientific evidence (double blind tests or
 GG> other scientific tests) that prove this is needed, I don't think this
 GG> is the way to improve digital sound.
I am not trying to say that wordlength and sampling rate are the only
problems that are worthy of attention, merely that they could both be
improved upon. Take care.
... Posted by the committee to bronze John Meyer.
--- Blue Wave/DOS v2.20
---------------