OOF Newsletter #6  August 2000

Contents:
   New Versions and Source Code Available
   Workshop
   Changes to OOF and PPM2OOF
   Mailing List Procedural Changes


Dear Colleagues --

New versions of OOF and PPM2OOF are available on the OOF web page,
         http://www.ctcms.nist.gov/oof/download/.
There you will find precompiled binaries for SGI, Sun (Solaris2), PC
(Linux), and Alpha (Linux).  The source code is also available,
for those who feel the need to run OOF on other architectures.
As always, please let us know if you encounter any difficulties.

We are considering having another OOF workshop at NIST in the coming
year. The workshop would give users a chance to learn about the
program and to exchange ideas on ways to use it, and would provide
valuable feedback to the developers.  Please let us know if you'd be
interested in attending.

The extensive lists of changes since the last newsletter can be found at
 http://www.ctcms.nist.gov/oof/download/oofchanges.html (for OOF)
and at
 http://www.ctcms.nist.gov/oof/download/ppm2oofchanges.html (for PPM2OOF).

Major changes to OOF since the last newsletter (June 1999) include:
 * Trigonal elements.
 * Tetragonal elements.
 * A Cross Section dashboard, for plotting stress, etc, along
   arbitrary lines in the xy plane.
 * An enhancement variable in the graphics menu, controlling the
   degree of distortion displayed.
 * Output on an element by element basis, but restricted to a single
   element group.
 * Warnings about non-positive element areas during equilibration.
 * Numerous bug fixes.

Major changes to PPM2OOF since the last newsletter include:
 * Configuration files, which store the current state of ppm2oof, so
   you can resume a session without having to re-run a log file.  The
   file saved by the "quit" command is now a configuration file, not a
   log file. Both configuration files and log files can be loaded with
   the -start command line option.
 * Text mode, for running ppm2oof without an X windows connection.
   Invoke text mode with the -text option on the command line.  All
   commands that don't require a mouse click position are available in
   text mode.  Commands that *do* require a mouse click will probably
   crash the program.  To find the text equivalent of a graphics
   window command, execute the command in a graphics window, and
   examine the log file (with the "log/save" command).
 * Active regions, allowing you to restrict most mesh and pixel
   operations to a subset of the image.
 * A more robust mesh refinement method, avoiding infinite loops that
   could occur in the old method in some (hopefully) rare circumstances.
 * Laplacian mesh smoothing via the "adaptive_mesh/smooth" command.
 * Undo and redo commands for mesh modifications, in the
   "adaptive_mesh/stack" menu.
 * New adaptive mesh refinement commands and parameters:
       refine interface (refine triangles with dissimilar neighbors)
       max_divisions (limit the number of times a triangle will be refined)
       min_area (don't refine triangles that are too small)
       division (criterion for deciding which way to divide a triangle)
 * A new Automate dashboard, which automatically selects pixels by
   burning random points on the image. It assigns the selected pixels
   to one of two groups, depending on the shape (aspect ratio and
   perimeter) of the randomly burned regions.  The dashboard was
   designed to distinguish between grain and grain boundary phases in
   one particular set of micrographs, but we hope it will be more
   generally useful.  (Thanks to Daniel Vlacich.)
 * A new Equalize command on the Image Modification dashboard, which
   equalizes nonuniform illumination.  That is, it ensures that the
   average brightness of a window around every pixel is the same as
   the average brightness of the whole image. (Thanks again to Daniel.)
 * Translucent colors for marking selected pixels, making it easier to
   see what's been selected.  The opacity can be adjusted in the
   graphics menu.
 * A -quit command line option, forcing ppm2oof to quit quietly after
   running a script.
 * Numerous bug fixes.


Finally, because of some recent spam, we've changed the mailing list so
that it won't accept messages from addresses that aren't subscribed to
the list.  If that causes a problem for anybody, send a message to
stephen.langer@nist.gov.

 The OOF team,
    Steve Langer
    Ed Fuller
    Andy Roosen
    Craig Carter

    oof_manager@ctcms.nist.gov