25 Sep
2009
25 Sep
'09
10:25 p.m.
On Fri, Sep 25, 2009 at 2:48 PM, Fons Adriaensen <fons(a)kokkinizita.net> wrote:
To keep things simple I've assumed constant
relative
speed errors. The situation is just the same if there
is some variation (wobbling) during the length of the
track.
i don't see how. on the "transfer out" step, you might get 32 minor
variations in speed, adding up to a total of 0.01 seconds of increased
length. during the "transfer back" step, you might get 28 variations
in speed, adding up a total of 0.008 seconds of decreased length. you
don't know the total variation till its done, and you don't even know
if it will be an increase or a decrease. if you "recorded" the 0.01
seconds of increased length, then any playback that doesn't precisely
match the error will end up being incorrect.
i'm missing something - this seems so obvious to me that i can't
believe you don't see this.
--p