Subject: Frequency response was Re: [linux-audio-user] Audiophile CD's
From: Jason (hormonex_AT_yankthechain.com)
Date: Sun Jan 27 2002 - 07:29:18 EET
> So I have number of questions:
> 1) Can regular CD/DVD drives on a PC read them? (If the answer here is no
> then the rest is irrelevant).
DVDROM drives SHOULD be able to read SuperAudioCDs. I say should because I
kind of doubt that the ones that exist in the real world actually do.
SACD's if memory serves are 20bit 88.2 kHz. Which is inferior to
DVDAudio's 24bit96k, so having a DVD laser read them should just be a
software issue. I think.
> 2) If so, then is there any Linux software to read and play them?
no clue. it shouldn't be too hard though, I imagine.
> 3) What about writing? (Surprising as it may sound, the regular CD format
> is actually not quite adequate for fully representing the sound on 78's,
> I've even seen signal [and I really do mean SIGNAL not the spectrum of a
> click] going up to 29kHz on an _acoustic_ disk [probably a resonance in
> the horn]).
The human ear can hear fundamentals roughly between 20Hz and 20kHz. It is
this poorly understood fact that led to CD's sampling at 44.1kHz.
According to the nyquist theorem in order to accurately represent a simple
signal
digitally, you have to sample it at twice it's own frequency, the result
of sampling at less than that frequency is a phenomenon known as Aliasing,
in which the samples actually give tehm impression of a mathematically
related lower frequency. It actually sounds kind of cool as an effect.
So anyway, A/D converters put these brick wall filters in line to
eliminate the frquencies above 22.05kHz. Since they have to be cheap in
most cases, they start doing that cuttoff a lot sooner. hence why sound
blaster and similar sound cards aren't really suitable for quality
recordings. cheap card = cheap converters = cheap filters = sound
attenuation above 15 kHz.
Any way, It's not really fair to compare CD's to vinyls because vinyl
doesn't have a flat frequncy response; it's an intentional distortion in
order to allow for higher dynamic range. In fact, a lot of the alleged
"warmth" of vinyl doesn't come from the vinyl at all, but rather the built
in compensation in the phono preamps. That having been said, vinyl is
allegedly supposed to be capable of accurate reproduction up to 25kHz,
although I don't have the kind of dog ears necessary to be able to comment
intelligently on that. However, I defy anyone short of geoff emerick or
carl beatty to point me to a sound system that isn't accurately
reproducing frequencies above 18k. It's really an issue that too much is
made of, IMHO, because most people can't hear it, and of the ones who can,
it's such a small portion of the available information, that it's
practically irrelevant.
What I do know, is that it is possible to hear into a much higher spectrum
than 20k when it comes to harmonics, and this is the one argument That I
acutally feel is valid when it comes to the dismissal of 44.1 as adequate.
Most people who can still hear 20k can distinguish between a 20k sine wave
and a 20k square wave. which means that they are at least hearing a few of
those overtones that you wouldn't be able to hear as a fundamental.
however, I do think that a bit too much has been made of the issue,
because as everyone knows, there are only two octaves between 20k and 80
k, so there isn't MUCH useful information there, so while 96k recordings
do sound nice, I really personally don't feel that it's nice enough to go
replacing my CD collection once DVD audio comes down the pike in a big
way. Good mixes and good converters mean much more than higher sampling
rates.
In other words, I wouldn't worry about buring SACD's, and I'd spend my
time and money on some really good EQ's, learning the exact implementation
of the RIAA reproduce curve in whatever phone preamp you're using,
buying fantastic apogee A/D converters and making really good standard
CD's. I think you'll find that reduced noise, the more even reproduction,
and the flatter response will be much more satisfying sonically that
simply upping the sampling rate ante.
somebody take away my soapbox, please...
~jason
-- YankTheChain.com - You can pretend we're not here. That's what I do.,
This archive was generated by hypermail 2b28 : Sun Jan 27 2002 - 07:20:27 EET