source file: mills2.txt Date: Mon, 17 Mar 1997 10:57:34 -0800 Subject: Re: unheard frequencies From: alves@Orion.AC.HMC.Edu (Bill Alves) Ray Tomes wrote: >It was discovered not so long ago that although people could not hear >frequencies above 22kHz which is the maximum CD frequency that some >could tell the difference between (say) 15kHz waves with different >timbre or different overtones. Some CD manufacturers now add overtones >to the high frequencies because that normally produces a more natural >sound. This little trick does fool those individuals who could tell the >difference before. I'm afraid I'm not following. How could CD manufacturers add overtones? If you mean that they are adding harmonics to the recorded sound, then the highest pitch that they could add a harmonic to would be 11 kHz, since its second harmonic (first "overtone") would be at 22kHz. All CD players include an anti-imaging filter that effectively eliminates all frequencies higher than 22k anyway (often with significant attenuation above about 18k). Some harmonics do occur because of distortion created in the analog circuits as well as the "squaring-off" of waves close to the digital resolution in either amplitude or frequency. However, audio companies work very hard to MINIMIZE these harmonics, as listeners find them very objectionable. Any digital harmonic distortion of tones above 11kHz should be eliminated by the anti-imaging filter. Harmonic distortion of waves close to the lowest representable amplitude is generally "smoothed out" by the introduction of dither, a small amount of noise, which effectively distributes the energy within the spurious harmonics throughout the frequency spectrum. Listener tests have shown that the extra noise is much less objectionable than the equivalent amount of harmonic distortion. It is very clear that very few people can consciously detect sine waves at 20kHz. Many people, in fact, have a hard time hearing above 15 or even 12kHz, depending on their age, sex, and lifetime exposure to loud sounds. I have heard that some people can detect timbral differences in very high sounds, but I have not yet seen a reference. I do not deny the possibility that ultrasonic sounds can have some kind of significant physiological effect, but I've yet to see any evidence to support that, much less that this effect is somehow perceptible. Bill ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ Bill Alves email: alves@hmc.edu ^ ^ Harvey Mudd College URL: http://www2.hmc.edu/~alves/ ^ ^ 301 E. Twelfth St. (909)607-4170 (office) ^ ^ Claremont CA 91711 USA (909)621-8360 (fax) ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Received: from ns.ezh.nl [137.174.112.59] by vbv40.ezh.nl with SMTP-OpenVMS via TCP/IP; Mon, 17 Mar 1997 20:15 +0100 Received: by ns.ezh.nl; (5.65v3.2/1.3/10May95) id AA03590; Mon, 17 Mar 1997 20:15:15 +0100 Received: from ella.mills.edu by ns (smtpxd); id XA03641 Received: from by ella.mills.edu via SMTP (940816.SGI.8.6.9/930416.SGI) id LAA12277; Mon, 17 Mar 1997 11:13:31 -0800 Date: Mon, 17 Mar 1997 11:13:31 -0800 Message-Id: <332D9DB5.6276@dnvr.uswest.net> Errors-To: madole@mills.edu Reply-To: tuning@ella.mills.edu Originator: tuning@eartha.mills.edu Sender: tuning@ella.mills.edu