[Gmsh] New Gmsh release: 1.53

Christophe Geuzaine geuzaine at acm.caltech.edu
Fri Jun 4 21:15:17 CEST 2004


I've just released Gmsh 1.53. This version contains fairly big changes,
so please report any bugs that might have crept in... The three largest
modifications are:

- big (huge?) improvements in the way Gmsh renders tri/quad meshes and
scalar post-processing views, with speedups of more than an order of
magnitude on large data sets (and as a side effect, transparency is now
fully supported)

- new geometrical entity selection with undo capability

- full support for second order elements in the mesh module (3-node
lines, 6-node triangles, 9-node quads, 10-node tets, 27-node hexas,
18-node prisms and 14-node pyramids)

See below for a summary of the other additions and bug fixes.

Downloads, mailing lists, etc.: http://www.geuz.org/gmsh/


New in 1.53: completed support for second order elements in the mesh
module (lines, triangles, quadrangles, tetrahedra, hexahedra, prisms
and pyramids); various background mesh fixes and enhancements; major
performance improvements in mesh and post-processing drawing routines
(OpenGL vertex arrays for tri/quads); new Plugin(Evaluate) to evaluate
arbitrary expressions on post-processing views; generalized
Plugin(Extract) to handle any combination of components; generalized
"Coherence" to handle transfinite surface/volume attributes; plugin
options can now be set in the option file (like all other options);
added "undo" capability during geometry creation; rewrote the contour
guessing routines so that entities can be selected in an arbitrary
order; Mac users can now double click on geo/msh/pos files in the
Finder to launch Gmsh; removed support for fltk 1.0; rewrote most of
the code related to quadrangles; fixed 2d elliptic algorithm; removed
all OpenGL display list code and options; fixed light positioning; new
BoundingBox command to set the bounding box explicitly; added support
for inexpensive "fake" transparency mode; many code cleanups;


-- 
Christophe Geuzaine
Applied and Computational Mathematics, Caltech
geuzaine at acm.caltech.edu - http://geuz.org