General

This program creates an artificial crystal array with specified degrees of ordering, randomness, and clustering at various scales, and saves a file that can be read into Reduce3D. It is designed for testing the algorithms of Reduce3D to ensure they produce correct results.

Memory Required: 512K (not stress-tested)

Input File Type

n/a

Output File Types

TEXT

Output Format

The file format is the same as that produced by Integrate and Crystallize, and can be read directly into Reduce3D.

ResTypes (not registered with Apple)

Application (Blob 1.0) = ‘qtMS’

Operator Notes

1. Most settings are self-explanatory with these exceptions:
   1. Use IC Overlap criterion - Disallow overlaps that would be impossible in an interface-controlled nucleation and growth regime (in which radial growth is proportional to time).
   2. Ordering uses popup - HCP is Hexagonal Closest Packing, the most ordered array possible; CCP is Cubic Close Packing, but really it should be SC, for Simple Cubic packing - the crystals are at the intersection of a 3-D gridwork.
   3. Nudge by - Moves each crystal by a very small location. This is to avoid a really difficult bug in the convex hull bounding box code of Reduce3D which blows up when lots of crystals share a single plane perfectly.
   4. Add random crystals - Adds random crystals into the ordered or clustered array to obscure the trends. The new crystals still have to obey whatever overlap criterion is set.

Program Notes

None at this time - it’s pretty simple code.

Known Bugs

None at this time.

Planned Improvements

None at this time.

Release History

1.0.0 - 4 April 99

Initial release. Well, actually, I never thought it would grow from a hacked-together utility to a nice program, so I never gave it a version number.

Authorship

GraphCFs was written by David Hirsch in early 1999.