[Gmsh] Medit file format

Geordie McBain gdmcbain at freeshell.org
Thu May 6 01:53:37 CEST 2010

2010/5/5 Christophe Geuzaine <cgeuzaine at ulg.ac.be>:
> Indeed, we do not read/write prisms or hexas in MEDIT format. If you know
> the correct vertex ordering in that format please let us know: we'll be
> happy to extend the routines to take thse kinds of elements into account.

Yes, I see the issue.  The document I take to be the specification for
MEDIT is Section 7.2.1 `The mesh format' of

Frey, P. J. (2001). MEDIT: An interactive mesh visualization software.
Technical Report 253, INRIA, Rocquencourt.

but it does not specify the vertex ordering; obviously this is
unsatisfactory, and I presume all other programs using the MEDIT .mesh
format will make some assumption for each element.  For example,
FreeFem++ agrees with Gmsh for the elements it uses: Edges, Triangles,
and Tetrahedra.

Actually, Frey (2001, Section 7.2.1) only defines Vertices, Edges,
Triangles, Quadrilaterals, Tetrahedra, and Hexaedra [so spelled];
prisms aren't mentioned.  Are there any applications out there
anticipating MEDIT prisms?

There is another document (in French):


but I don't think it gives the vertex ordering either (though my
French is very limited).  It doesn't have prisms.

I'll write Pascal Frey and ask him about the vertex ordering of
hexahedral elements.