version 15.76

The update to version 15.76 incorporates the following changes:

Features:

  1. The logic of the PAD scan (which shows the ray curves as the field is scanned from the axis to the edge) has been modified.  Previously, the program worked but suffered from the tendency of Windows to deactivate the view after a pause since there is then no user input.  Then the screen would freeze until the scan was finished.  The new logic seems not to suffer for that reason.  The Zoom Slider has similarly been enhanced.
  2. SFIT has been enhanced.  Now it can be used to augment an existing aspheric surface with measured figure errors, as well as fitting the whole surface to new input data.
  3. A new feature, LNC will list the NURBES control points in X, Y, and Z.  The ASY listing gives the Z, but only implies X and Y.

Bugs Fixed:

  1. A problem was reported with user-written DLL files.  This has been resolved, and all users who need this feature are advised to contact OSD for an updated header file that is needed by the program.
  2. The DSEARCH and ZSEARCH programs did not work properly in multicore mode after a recent update.
  3. The option in DSEARCH to allow up to four trial radii only worked with three or fewer.

 

version 15.77

The update to version 15.77 incorporates the following changes:

Features:

  1.  LNC will now list both the NURBS control points and the surface sag at each point, for reference.
  2.  AAA (Automatic Asphere Assignment) now has an optional EXCEPT… entry.  This will exclude surfaces that you don’t want to become aspheric.  ADA (Automatic DOE assignment) is similarly enhanced.
  3.  If you ask for the monitor AGE or AAE after you have already declared AEC, instead of dropping the latter, the program now converts it to the other unused type.  So you won’t wonder why your lenses are feathered anymore.
  4.  Several new glass types have been added to the Schott, Ohara, Hoya, and LZOS catalogs.

Bugs Fixed:

  1.  The ZMC (Zemax conversion) rendered YFLN data as OBA,  It should have been OBC.
  2.  ADA in multicore mode would crash if switch 99 was turned on.  Since ADA already runs multiple processes, it was not able to run each one in another set of multiple processes.  Now it will turn off switch 99 first.  The same was true for the Monte-Carlo program MC.
  3.  The REVERSE command did not reverse the aspheric coefficients on a USS 16 DOE.
  4.  The lens layout tool LLT, when run on a lens with object type OBD, did not work correctly unless the system was in GLOBAL mode.  Now object OBD turns on that mode automatically.

 

version 15.78

The update to version 15.78 incorporates the following changes:

Features:

  1. The step file program STFILE now can model all axially-symmetric aspheric surfaces with the native NURBS specification in that protocol, so those won’t be rendered as panels anymore.  (The panels were accurate enough but created very large files.)  Non-symmetric shapes or those without circular and centered apertures are still rendered as panels.
  2. A new feature is available in FILM:  Now you can declare a custom library of film materials by entering the six interpolation coefficients in either the old Schott or Sellmeier formulas.  Then you can define a custom coating stack that incorporates those materials and apply it to your lenses. Section 16.5 in the UM describes this new feature.
  3. AED will now check whether an element it wants to remove from the lens is at the aperture stop.  If so, it will replace it with a dummy surface and keep the stop there.  Before, it simply deleted the element, which forced the stop to move to the previous one, which was sometimes not a good idea.
  4. The structure of the header file fortranstructures.h has been altered.  Users who have created a custom DLL file should contact OSD to get the latest version.

Bugs Fixed:

  1. The option in ZSEARCH to space the zoom positions with forms LINEAR and NONLINEAR did not preserve the Lagrange invariant.
  2. The GDF option to input glass interpolation coefficients did not work properly.
  3. The DMASK listing of the blaze height was not correct on the plot.
  4. The option in FILM to fix the layer material to THF4 did not work properly if the wavelength exceeded the visible-spectrum range of that material.  It then defaulted to the IR range, which was not accurate near the visible.
  5. Also in FILM, the analysis of coatings accounted for the wavelength and angle of incidence, but ignored the dispersion of the layer material, which has a small effect.
  6. DSEARCH did not report the number of cases it evaluated correctly if the number of authorized cores exceeded the total number of cases.
  7. The option in ZSEARCH to automatically correct ray failures before optimizing each test case did not work properly.

 

version 15.79

The update to version 15.79 incorporates the following changes:

Features:

  1.  The plastic material model varied by PGM has been improved and is now used more effectively.
  2.  A problem has emerged related to switch 99, which makes single-configuration optimization utilize multiple cores.  While this works as intended and greatly speeds things up, we have found that the Windows operating system often fails to work properly afterward.  We are researching this problem and have provided a new command, C99, that controls how many cores are used for this purpose.  Fewer used seems to reduce the incidence of this effect.  Users are advised to experiment, and let us know if you gain any insight into what may be causing this strange effect.

Bugs Fixed:

  1.  The number of cases listed as skipped by AEI was sometimes incorrect.
  2.  GSEARCH did not work properly if switch 99 was turned on.  Now it turns that switch off before it runs.
  3.  The Free-form system build, FFB, when a circular field was specified by CFOV, created the correct set of rays for the merit function, but did not assign the CFOV flag to the system itself.
  4.  The convolution MTF became inaccurate if the lens pupil was strongly noncircular.
  5.  If you ran DSEARCH with multiple trial radii values, and the number of authorized cores exceeded the number of cases given by the binary search algorithm, some cases were examined more than once, which could potentially eliminate other good solutions.

 

 

You can download this version at the following link:

http://www.osdoptics.com/installation.php

If your support is current, you can run the full version of this release. All users are encouraged to keep their support current and to install updates as soon as they are available.