To: "wgoldys"
,JulesFran{at}aol.com,atm{at}shore.net
From: Jim Burrows 
Reply-To: Jim Burrows 


1)  FAST CONICS - Walt Goldys asked a week ago, "Why isn't Sixtests'
sphere deviation plot flat when the input data is from a sphere (constant
Foucault readings)?"  This has been fixed.  The fix was adding two
more Zernike coefficients to the 3 I was using to to find the best-fit
conic (the minimum number), which works for all parabolas, but not for
non-parabolic conics with f/ < 2 or 3, and using the the exact conic
formula for plotting and calculating RMS.  The deviation plot doesn't look
flat for a 100mm f/2 sphere because of scaling, but the error is less than
1 nm.  Also a small change to the help file that Walt suggested.

2)  INCHES - several users have wanted inches instead of those gol' dang
millimeters (no Europeans ).  I've gone halfway - if you put
"INCH" in as the last line of the input file, the program now
goes through the arduous task of multiplying all the input data by 25.4.

The program is otherwise unchanged, so those who don't need the above
changes won't have to download:

         http://home.earthlink.net/~burrjaw/public/win6.zip
(2003-01-26, Win32, 465k)

         -- Jim Burrows
         -- mailto://burrjaw{at}earthlink.net
         -- http://home.earthlink.net/~burrjaw
         -- Seattle N47.4723 W122.3662 (WGS84)

--- BBBS/NT v4.00 MP