To: atm{at}shore.net
From: jsgrmfg{at}gracemfg.com
Reply-To: jsgrmfg{at}gracemfg.com


>Dave Rowe and I are working on porting (actually re-writing) Figure45 over
>to a Win32 environment.

   Goody.  I like Figure45.  I recently got a new computer running
Win XP and my Figure45 no longer works :-( .

>#1 Does anyone actually "use" any of the following in Figure45
   > A) Transverse error analysis graph
   > B) Spot Diffraction Analysis graph
   > C) Spot Visualization Display

   Nope. Never use 'em.

    >More importantly, would anyone severely miss these options if they
were
>no longer available?  Personally, in our ATM lab classroom environment, we
>never use these options.  While keeping these options isn't too much work,
I
>thought I might instead concentrate on adding new options like:

    >A) A figuring session comment screen
    >B) Figuring history slide show (similar to Couder.exe by Ricardo
Dunna)
    >C) Find Best fit Conic Constant screen (Similar to Sixtests by James
>Burrows)
    >D) New surface analysis screen that:
        >i) Excludes central obstruction
        >ii) Allows masking off the edge of the mirror

   Those all sound like useful additions to me.

>#2 Surface Error Analysis Graph Screen

>Have a look at this, and tell me what you think:
>http://lerch.no-ip.com/atm/FigureXP.gif ( 24KB)

   Looks nice.  My only objection to a full diameter graph is that
it necessarily reduces the vertical scale of the graph, making it a bit
harder to read.

>Here's the proposed changes:

    >A) While I was writing the code for this display, I got to thinking
>about our classroom environment.  Nearly every time I show the classic
>radial surface deviation graph to a new student, their eyes glaze over.
So
>my first thought was to chart the surface error over the Diameter of the
>mirror, instead of just the radius.

    >B) X axis labels.  Instead of  radial percent, I just displayed the
zone
>center radiuses.  Compare the graph to your couder mask (or rings drawn on
>the back of the mirror blank) to help visualize the areas that need work
>while figuring (Does anyone want tick marks showing radial % or physical
>dimensions?)

   I like the sixtests feature of tick marks at the pin or mask opening
locations.  They are sometimes very helpful in interpreting a graph.

    >C) Y axis labels.  Instead of marking off microns (or nanometers) I
just
>put wave error reference lines.  (Does anyone really want to know the
actual
>surface height in nanometers??)

   I assume that is surface error, not wavefront error.  It would be easy
to get the two confused.  Perhaps a prominent NOTE somewhere on the page
would keep things straight.  A Y axis marked in nanometers would make it
easier to compare results with Sixtests.

Jim Sturtevant

--- BBBS/NT v4.00 MP