[Gmsh] Possibly a bug?

[Geordie McBain]

2011/10/27 rakesh patil <prakeshofficial at gmail.com>:
> Hello,
>
> I noticed one thing in gmsh. I used same single .geo file in multiple
> instances of gmsh and tried to generate a mesh. And everytime the number of
> nodes and elements were different for same .geo file. Is this a bug or is
> that how it should behave?

This is expected behaviour; see e.g.

  http://www.geuz.org/pipermail/gmsh/2011/006431.html

Lots of unstructured meshing algorithms contain random steps; see e.g.

Persson, P.-O. and G. Strang (2004). SIAM Review  46 (2), 329+.
http://dx.doi.org/10.1137/S0036144503429121