John,
To challenge this statement of yours : I once tested my then new Marantz CD-63 MkII KI Signature CD-player (a respectable one in my opinion) and played a CD where I had put a 20 kHz sinus testsignal on (made with CoolEdit, so with mathematical precision). The output signal was a perfect distortion free sinus of the exact amplitude which proved the validity of the Nyquist sampling theorem to me. Don't let your mind fool you (it indeed sounds strange that 2 samples per period suffice to reproduce a periodical sinewave, but I assure you : it is) !
Kind regards,
Luc Henderieckx luc.henderieckx@pandora.be http://users.pandora.be/airborne
It's really hard to explain why upsampling works without resorting to graphs. Basically, audio theory says that 40,000khz is adequate to play back a 20khz sinewave. But if you think about it, a sampling rate of only 44,000 hz gives a VERY crude approximation of a sine wave. Sure, the
peaks
and the valleys are there, but the sine wave now looks more like a square wave than a sine wave.
With upsampling, you can interpolate that waveform to *approximate* a sine wave. Obviously, the best solution would be to do the original recording
at
a extremely high sampling rate.