Re: [ATM] Touchy Refractive Index

[Woodchuck <djv@bedford.net>]

On Sat, 15 May 2004, Bob May wrote:

> I've come to find that your findings are quite real as I've seen a lot of
> changes in the spot diagrams and so forth when playing with glasses in the
> several programs that I've been using for refractor design.
> I've come to believe, without having built any of the designs, that once you
> get down near the final results, you pretty much have a good design with
> anything near the right numbers.
> Finally, I'd find that actually building any of the designs will find that
> production errors in thickness, real refractive indices and actual ROC of
> the curves will always make something a bit different than what the design
> was supposed to be which is why there is so much final futzing about with
> refractors in the end.
> Bob May
> http://nav.to/bobmay
> bobmay@nethere.com
> NEW! http://bobmay.astronomy.net

Completely off-the-wall Q: (I don't/can't run OSLO and haven't
gotten around to writing my own)   --  Would it make any sense to
run the optimizer with a cost function that includes sizeable
contributions from the various manufacturing sensitivities?
(i.e  d("image")/d(refractive-index), d("image")/d(radius[i]))

In old books it is said that the variation in refractive index
in a "real" glass of "real" size is always large enough to swamp
minor errors of radius, and that not only is figuring always
anticipated, i.e. emperically aspherizing the final surface, but
also secondary retouching of local areas.  Is this still true?
Was it always true?

Dave
-- 
  Who would've ever thought that more nude pictures would have come
  out under President Bush than under President Clinton?
                           -- Jay Leno

_______________________________________________
ATM mailing list http://www.atmlist.net/