RE: [ATM] Round robin mirror C Foucault profiles

["Koehler, Steve" <Steve_Koehler@securecomputing.com>]

I tried another slant at comparing surface profiles from the mirror C test data.
Dale sent me his averaged Sixtests files (except that Scott R. was missing), and
I used Mike Peck's modal method of deriving the surface.  One other thing I did
differently than Dale was to use the same zone radii and ROC for all test
results, rather than using the numbers reported by the testers.  I think this
puts the tests on a more even footing.

The results are in the directory:

http://www.atmlist.net/contrib/steve_koehler-at-securecomputing-dot-com/roundrob
in/

   c_modal_3.png fits three symmetric zernike terms
   c_modal_4.png fits four symmetric zernike terms
   c_modal_5.png fits five symmetric zernike terms

There is not enough data to go above five terms.  The 5-term plot shows a
remarkable similarity between the seven reported testers.  I would ignore what
the curves do inside the center of zone 1 and outside the center of zone 5
(shown on the plots), because this is an extrapolation with little data.  Note
that the high and low points shift a bit with the 3- and 4-term plots.  This is
because the reduced set of symmetric zernike terms can't make curves with the
humps in the right place.

-- Steve Koehler
   steve_koehler@securecomputing.com


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