Justin,
You're correct, 100 cents grows depending on register. The cps width of =
100
cents between middle-C and middle-C# is (261.626 cps) to (277.183), but =
the
cps width of 100 cents between B above middle C and the 1/2 step above =
that
(C-octave above) is 493.883 to 523.251.
The other issue has to do with how often you sample the fundamental
frequency of a live-performed pitch. I don't have time, but in Csound I =
can
sample a fundamental at a granularity of 400cps. When one does that, =
one
finds that live performed pitches vary in pitch (up/down as much as 20
cents) even within the span of one second - especially in performance
traditions that incorporate vibrato and timbre changes - and this is =
where I
question the validity of the test that produce statements like "tests =
prove
that people can ID pitch-relations within a 5 cent tolerance." It seems =
to
me that in cases like that, those who are testing the responders must =
simply
be taking them at their word.=20
Dan Gutwein
Instructional Multimedia Specialist
HR / Center for Professional & Organizational Development
Office of Information Technology Building
OITB, N3 (240) 314-3159
_____________________________________________
List un/subscription information is at:
http://communication.ucsd.edu/shaku/listsub.html
This archive was generated by hypermail 2b29 : Fri Jan 06 2006 - 10:00:47 PST