Subject: Re: Frequency response was Re: [linux-audio-user] Audiophile CD's
From: Ross Vandegrift (ross_AT_willow.seitz.com)
Date: Sun Jan 27 2002 - 22:00:23 EET
> I believe the nyquist theorem is correct as far as it goes but did not take
> into account that the sounds created by an instrument that are outside of our
> hearing range.
Um... the Nuquist Theorem doesn't really have anything to do with sounds, in
particular. It's something that is true for *all* waveforms. It doesn't need
to take into specific account freqencies we can't hear - I can apply it to light
waves and it still works.
Weisstein says in the CRC publishing of his website (which I purchased before
they sued him for the site...) under the entry for the Sampling Theorem that to
reproduce all of the Fourier frequencies for a waveform, one must sample at a
frequency f, where f >= 2B. Sampling at f = 2B is called Nyquist Sampling.
Sampling at f > 2B is called Oversampling, and does not add artifacts or lose
any information.
A quick Googling for "nyquist theorem proof" turns up few precise proofs, but
also more mathematical analysis of Shannon-Nyquist than I could shake a stick
at.
Twice the bandwidth - not just a good idea: It's the law.
> I haven't
> heard of too many tools being created to measure what we don't hear and how
> it's persistance in existing affects what we can hear.
Well, if a sound outside of my hearing range affects a sound inside my hearing
range, I don't need to study the sound that was too high - if I could hear
artifacts of the interaction, I can simply study the interaction.
> (you try playing a trumpet quietly (chuckle) but it don't sound much like a
> real trumpet).
As a student of classical trumpet, I resent the implication present here! ::-)
Ross Vandegrift
ross_AT_willow.seitz.com
This archive was generated by hypermail 2b28 : Sun Jan 27 2002 - 21:53:41 EET