edit The noticeable change seems to be very close to 22s.
edit The noticeable change seems to be very close to 22s.
I'm not quite sure what I imagined the experience would be, but to my perception you're right... even when the switch-over (from the higher note as the most prominent, to the one an octave below) is very smooth, there's just a moment where you hear both notes clearly, then the higher note fades out.
It might be more interesting to play with Shepard tones in a more complicated arrangement -- i.e., instead of one prominent note that fades into the next one an octave down, you might have 4-5 prominent notes, fading into others with staggered switch-overs -- possibly all at octave splits, but possibly in a more interesting harmonic blend.
The idea is still interesting. :)
If you don't correct for that when synthesizing your tones, then one pitch may be much more prominent than the others.
I'm curious if you happen to be a musician or a sound engineer?
That falling tone on the Wikipedia page sounds like something from one of coil's tracks. I can't pinpoint which one though? they use similar sounds here http://www.youtube.com/watch?v=c6IR90tgtDI and here http://www.youtube.com/watch?v=9_R-mXRONh0
edit Expectation may pay a part - when told "come and listen to this 'endlessly falling' auditory illusion" of course I'm going to try and pick out the subtle seams. I might not notice if it was in the background, unannounced.