Re: [ATM] FigureXP update

[Michael Peck <mpeck1@ix.netcom.com>]

At 01:08 16-07-04, James Lerch wrote:
>Greetings All,
>
>After reading all the discussions on the Millies-Lacroix criteria, I thought I
>would include this method into FigureXP.  Hopefully this will help spawn
>discussion on the merits of various evaluation methods, as they can now be 
>more
>readily compared.

James:

I'm not entirely sure why you want both an M-L and a transverse aberration 
analysis in Figure since they tell you essentially the same thing. But if 
you really want them both they should produce consistent results, which 
they don't quite do yet.

I entered a 150mm f/9.56 sphere, which I claimed a couple weeks ago just 
satisfies M-L. I "tested" it on 6 zones, letting your mask wizard set up 
the zone radii, and entering 0 for all the zonal readings.

The surface error plot correctly shows a Strehl of about 0.92 and surface 
RMS=12.7 nm = .046 waves on the wavefront. Going to the M-L plot and 
clicking away all the points eventually fall within the tornado with a 
reported maxRTA of 0.9. It should be 1, but maybe that's because of not 
enough zones in the "test". Going next to the Transverse error plot and 
again clicking away I get a minimum RTA of 1.202. I think the 1.2 is 
because Dave's original version of Figure divided transverse error by 
lambda*f/D instead of 1.22*lambda*f/D.

Three other comments:

1) The relative TA plot is wrong. It really should only go from 0 to the 
mirror semidiameter. But if you want to show the whole diameter the left 
half of the diagram should be the negative of the right half.

2) Adjusting the constant offset in a M-L analysis implies a choice of a 
reference surface, just as choosing a center of curvature offset does. Your 
Transverse error plot window correctly updates the surface error summary 
information as you click on the errors. The M-L plot window doesn't. You 
should fix this, which makes clear that improving "M-L" makes the surface 
error worse.

3) When you go from the Transverse error or M-L screen back to the surface 
error screen the surface error resets to the minimum RMS surface. I'd 
suggest *not* resetting it. This will make clear that optimizing according 
to RTA or M-L is not the same as optimizing RMS error.

Mike Peck


_________________

Michael Peck
email mpeck1@ix.netcom.com
Wildlife photography page http://home.netcom.com/~mpeck1/index.html
Amateur telescope making http://home.netcom.com/~mpeck1/astro/astro.html 

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