Twelve Ways to Fool the Masses When Describing Mesh Generation Performance

Proceedings, 13th International Meshing Roundtable, Williamsburg, VA, Sandia National Laboratories, SAND #2004-3765C, pp.181-190, September 19-22 2004

MESHING
RESEARCH
CORNER

13th International Meshing Roundtable
Willimasburg, Virginia, USA
September 19-22, 2004

Sandia National Laboratories, Albuquerque, NM
tjtautg@sandia.gov, drwhite@sandia.gov, leland@sandia.gov

Abstract
Mesh generation for finite element analysis is far from a solved problem. Although several automatic meshing algorithms exist, other difficulties of setting up a problem for finite element analysis still make this an interactive process. Undaunted, we continue to perform research on, and therefore publish, papers describing their work to overcome these problems. In our efforts to describe our work “in the best possible light”, we often obscure the real technical issues in these publications, rather than honestly assessing both the pros and cons of the described approach. This is an especially fruitful endeavor when reporting on mesh generation, given the 3D nature of the problem and the natural tendency of layman to avoid understanding the details of this problem. This paper describes twelve tried-and-true methods for obscuring rather than elucidating the performance of mesh generation technology.

Download Full Paper (PDF Format)