>>>>> "Tom" == Tom Jennings <tomj_at_wps.com> writes:
Tom> On Fri, 2004-09-24 at 12:06, Vintage Computer Festival wrote:
>> > http://www.fi.edu/pieces/knox/automaton
>>
>> Now THAT is amazingly cool.
Tom> Those things really are extraordinary. Even more amazing must be
Tom> the method of describing, in cams and levers, the motions
Tom> necessary to carry it out. It's "programmable".
Sure is.
Tom> I *never* hear any interest, discussion, or documentation of how
Tom> the process of 'programming' these machines (as well as music
Tom> players and related automata) was done. There's a transformation
Tom> and abstraction of thought, through arbitrary hardware
Tom> constraints, into a desired end. It's an algorithm, and (pun
Tom> intended) if it walks like a duck, it's probably a duck -- it's
Tom> programming.
In some cases there isn't much abstraction. For example, music boxes,
player pianos, and calliopes all have a tape or roll with one
"channel" per sound making element, moving at constant speed. So the
program is simply a two-dimensional representation of the sound:
frequency (or more precisely, sound actuator) on the x axis, time on
the y axis.
That cam mechanism is similar to what you find in many mechanisms,
though a whole lot more complicated. Describing motion of mechanisms
by one or more cams is classic mechanical engineering. Design of
cams, in particular cams in fast moving mechanism, is a serious topic
for ME majors...
paul
Received on Fri Sep 24 2004 - 15:44:10 BST