Re: "analog computer" = useless hypothesis?

From: Dave Sill (extropians@dave.sill.org)
Date: Wed Apr 04 2001 - 20:13:20 MDT


"J. R. Molloy" <jr@shasta.com> wrote:
>
> For computing purposes, can we then conclude that "analog" machines have no
> use?

No, definitely not. There are special-purpose analog computers in use all
over the place. Woodworkers use a simple device consisting of a right angle
with straight edge at 45 degrees to quickly and accurately locate the center
point of round stock without measurement. In the days before CAD, draftsmen
used a simple technique to evenly subdivide a distance by angling a rule
until it precisely measured the desired number even units.

> Are these really "analog," and are they really "computers"?

The examples I gave are clearly analog, and do compute a result, though they
aren't computers in the programmable, general purpose sense.

> Is there really no such thing as a real analog computer?

One of my favorite examples is the "shortest path" problem. Given a 2- or
3-D map, with a list of nodes and arcs, determine the shortest path between
two specified nodes. Digitally, this is a pain in the ass. The analog
solution is trivial: build a scale model of the graph with string. Pinch the
two nodes, one with each hand. Pull hands apart. The distance between the
hands is the length of the shortest path.

-Dave



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:44 MDT