mirror of https://github.com/CGAL/cgal
paste some comments from Sebastien Valette
This commit is contained in:
parent
b920c18b73
commit
faa93e2894
|
|
@ -50,7 +50,18 @@
|
|||
#include <vector>
|
||||
|
||||
|
||||
|
||||
// Early convergence ratio which is used for the first two minimization steps.
|
||||
// The final convergence is carried out entirely i.e. convergence is reached only when no more optimization can be performed.
|
||||
#define CGAL_TO_QEM_MODIFICATION_THRESHOLD 1e-3
|
||||
// Used to clamp curvature values:
|
||||
// - regions with zero curvature (although this is not really a degenerate case....)
|
||||
// - regions with too high computed curvature (sometimes due to numerical issues in curvature computation)
|
||||
// Clamping happens this way ;
|
||||
// -The average weight weight_avg is computed
|
||||
// weights higher than CGAL_WEIGHT_CLAMP_RATIO_THRESHOLD * weight_avg are set to this value
|
||||
// weights lower than 1/( CGAL_WEIGHT_CLAMP_RATIO_THRESHOLD * weight_avg ) are set to this value
|
||||
//
|
||||
#define CGAL_WEIGHT_CLAMP_RATIO_THRESHOLD 10000
|
||||
|
||||
namespace CGAL {
|
||||
|
|
|
|||
Loading…
Reference in New Issue