[Gmsh] Different results for different stations or users

[RENARD Francois]

Hello,

Thank you for this great mesh tool !

I obtain (slightly) different results when running GMSH from the same 
geometry (.geo file) :

- on different (linux CentOS 6.7) stations

- on the same machine but from different users

For example two different users have obtained the following different 
results on the same station :

First user : 86048 vertices 553346 elements 382088 tetrahedra
Statistics:

   Input points: 50580
   Input facets: 101156
   Input segments: 151734
   Input holes: 0
   Input regions: 0

   Mesh points: 50580
   Mesh tetrahedra: 151938
   Mesh faces: 354454
   Mesh edges: 253095
   Mesh faces on exterior boundary: 101156
   Mesh edges on exterior boundary: 151734
   Mesh faces on input facets: 101156
   Mesh edges on input segments: 151734

Info    : 50580 points 253095 edges and 101156 faces in the initial mesh
Info    : Found region 1
Info    : 0 points created - Worst tet radius is 24.375 (PTS removed 0 0)
Info    : 4927 points created - Worst tet radius is 1.73689 (PTS removed 
0 73)
Info    : 9901 points created - Worst tet radius is 1.45257 (PTS removed 
0 99)
Info    : 14867 points created - Worst tet radius is 1.29636 (PTS 
removed 0 133)
Info    : 19826 points created - Worst tet radius is 1.1871 (PTS removed 
0 174)
Info    : 24786 points created - Worst tet radius is 1.10673 (PTS 
removed 0 214)
Info    : 29730 points created - Worst tet radius is 1.05069 (PTS 
removed 0 270)
Info    : 34664 points created - Worst tet radius is 1.00635 (PTS 
removed 0 336)
Info    : 3D point insertion terminated (86048 points created):
Info    :  - 344 Delaunay cavities modified for star shapeness
Info    :  - 340 points could not be inserted
Info    :  - 382088 tetrahedra created in 6.25905 sec. (61045 tets/sec.)
Info    : Done meshing 3D (14.4008 s)
Info    : 86048 vertices 553346 elements
Info    : Writing 'geom3d.msh'...
Info    : Done writing 'geom3d.msh'
Info    : Stopped on Fri Sep 18 15:47:55 2015

Second user : 86048 vertices 553388 elements 382130 tetrahedra
Statistics:

   Input points: 50580
   Input facets: 101156
   Input segments: 151734
   Input holes: 0
   Input regions: 0

   Mesh points: 50580
   Mesh tetrahedra: 151938
   Mesh faces: 354454
   Mesh edges: 253095
   Mesh faces on exterior boundary: 101156
   Mesh edges on exterior boundary: 151734
   Mesh faces on input facets: 101156
   Mesh edges on input segments: 151734

Info    : 50580 points 253095 edges and 101156 faces in the initial mesh
Info    : Found region 1
Info    : 0 points created - Worst tet radius is 24.375 (PTS removed 0 0)
Info    : 4930 points created - Worst tet radius is 1.73677 (PTS removed 
0 70)
Info    : 9906 points created - Worst tet radius is 1.45214 (PTS removed 
0 94)
Info    : 14871 points created - Worst tet radius is 1.29598 (PTS 
removed 0 129)
Info    : 19830 points created - Worst tet radius is 1.18694 (PTS 
removed 0 170)
Info    : 24789 points created - Worst tet radius is 1.10668 (PTS 
removed 0 211)
Info    : 29731 points created - Worst tet radius is 1.05068 (PTS 
removed 0 269)
Info    : 34670 points created - Worst tet radius is 1.00629 (PTS 
removed 0 330)
Info    : 3D point insertion terminated (86048 points created):
Info    :  - 345 Delaunay cavities modified for star shapeness
Info    :  - 335 points could not be inserted
Info    :  - 382130 tetrahedra created in 6.28704 sec. (60780 tets/sec.)
Info    : Done meshing 3D (14.4428 s)
Info    : 86048 vertices 553388 elements
Info    : Writing 'geom3d.msh'...
Info    : Done writing 'geom3d.msh'
Info    : Stopped on Fri Sep 18 14:26:18 2015

Is it possible to better control the gmsh calculations in order to 
obtain the same results ?

Thank you