From: "Scott Berfield" 
To: "Bill T." , 
Reply-To: "Scott Berfield" 


The cool thing about genetic/evolutionary algorithms is that while you can
count on getting something that solves your requirements, you have no idea
how it is going to work until you see it. In some cases, you don't know how
it works even after you see it. Unless you added constraints to point
toward spherical solutions, since the algorithms don't actually have to
push glass, I can pretty much guarantee you'd get all sorts of wild
surfaces interacting in totally non-intuitive ways -- but that would work
really well if you could build them.

One thing they have learned from this sort of thing is that the
selection/survival criteria need to be very carefully planned if you are
going to get what you want in the end. Once experiment used a system based
on an FPGA that could re-program itself. They ended up with circuits that
were incredibly efficient at solving their problems, but that only ran on
that specific chip and could not be transferred to any other device. The
algorithms had come up with efficiencies based on the specific structural,
chemical, and electrical idiosyncrasies of that specific piece of silicon
-- just like over specialization in nature.

It sure would be interesting to see this tried for optics.

-----Original Message-----
From: owner-atm{at}shore.net [mailto:owner-atm{at}shore.net] On Behalf Of Bill
T.
Sent: Wednesday, June 04, 2003 7:46 AM To: atm{at}shore.net
Subject: Re: ATM Darwin Optical Design


The truth comes out...now we know where Don's focuser design came from!
;^)

Looking at those strange and wonderful circuits in the Scientific American
article, it seems to me that those same programs turned loose on optics
would crank out anything BUT pure spherical-surface designs.

Bill T.


>In the Feb 2003 issue of Scientific American  the article of Evolving
>Inventions By John R. Koza, Martin A. Keane and Matthew J. Streeter
>presents an interesting concept of design by computer. Optimization
>programs found in some ray tracing software and electrical simulation
>(spice) work only on a range of values of a predefined optical or
>electrical design. The new type of software presented in this article
>actual mimics the process of evolution. The design itself, components
>type, component placement and range of values, are allowed to change
>and "evolve" and produce new combinations in the design. Some of the
>examples in the article are of analog filter circuits that were not
>only optimized, but, new and better circuit topologies emerged. The
>same could be applied to optical design...perhaps an
>all- spherical large easily built scope design might be realized.
>
>
>
>Don Clement
>
>Running Springs, California
>
>http://www.clementfocuser.com
>
>

_________________________________________________________________
The new MSN 8: smart spam protection and 2 months FREE*
http://join.msn.com/?page=features/junkmail

--- BBBS/NT v4.01 Flag-4