While on the thread of optically scanning musical box discs, it
occurred to me that one important point may so far have been
overlooked.

If one is using a "tracker bar" arrangement to do the reading should it
not be curved in the plane of the disc rather than straight ?

The surface velocity of the outer disc projections is much greater than
for those near the centre (hence the treble end of the comb is always
near the edge of the disc).  This causes the star wheels at the edge to
advance much more quickly than those near the centre resulting in a
relative delay in the bass that should be compensated.

Think of a taught wire hooked over the centre spindle and pulled
radially along the top edge of the star wheels .  If I use it as a
substitute for the disc projections to rotate the star wheels by
pulling it tight and moving it in the direction that disc would
normally rotate, would not the treble notes near the edge play before
the bass notes at the centre ?

I expect this situation could also be managed with a straight tracker
and software but I don't think Cakewalk is up to it.

Just my thoughts.

Jim in Sechelt