> Can I have one of your t-shirts?
All sold out. BTW, speaking of t-shirts, there is a lady in our church
that has a t-shirt business. Her highly successful ones are variants
of "I'm Iowish".
> This seems to me to be correct. [regarding scanner resolution vs image
> resolution]
The parallel to this is audio. To capture a 20kHz tone would require
double the sampling rate. However, that then doesn't differentiate
between what kind of audio wave it is: Square, sawtooth, sine--they
all capture identically. As the audio frequency gets closer to this
threshold, the more important oversampling becomes. Othewise Nyquest
kicks in awkwardly and the low-pass filter (anti-aliasing filter)
reduces not only frequency response (resolution), but alters the very
shape of the waveform.
To determine the shape of the wave, not just the frequency of the wave
requires far more samples than what the simple math would indicate.
To run this metaphor a little further, this also explains a little bit
why analog recordings sound different than digital recordings when
dealing with some audio sources like violins and woodwinds. The
waveforms in the overtones are extremely complex. Just capturing the
overtone isn't enough. Capturing the shape of the overtone is
important. Rupert Neve figured this out a few decades ago and
determined that the audio chain had to be transparant three octaves
above the threshold of hearing. A 96kHz sampling rate is only ONE
octave above the threshold of hearing.
AG Ears
--
Ken Norton
ken@xxxxxxxxxxx
http://www.zone-10.com
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|