Errata for Vega FEM v2.0

The following is a list of all known issues in Vega FEM v2.0. All issues are already resolved in the latest zip file available for download (and in version 2.1).

  • imageIO.cpp, line 511: change "uint32 tiff_bits" to "uint16 tiff_bits". Bug fix incorporated into the distribution on Oct 16, 2013; prior downloads must manually fix this.
  • PardisoSolver.cpp: ForwardSubstitution, DiagonalSubstitution, BackwardSubstitution: for correct results, must set iparm[7] to 0 prior to calling PARDISO. This issue does not affect the main Pardiso functionality in Vega, or any of the provided demos; it only affects the (esoteric) case where one wants to dissect the system solve into the three individual substitutions. Bug fix incorporated into the distribution on Oct 18, 2013; prior downloads must manually fix this.
  • homogeneousMooneyRivlinIsotropicMaterial.cpp: missing implementation of GetCompressionResistanceFactor. Incorporated into the distribution on Oct 27 2013; prior downloads must manually fix this.
    double HomogeneousMooneyRivlinIsotropicMaterial::GetCompressionResistanceFactor(int elementIndex)
{
  return EdivNuFactor;
}
  • objMeshGPUDeformer.cpp: must use OpenGL/glext.h on Apple platforms as opposed to GL/glext.h. Incorporated into the distribution on Oct 27 2013; prior downloads must manually fix this. 
    #if defined(WIN32) || defined(linux)
  #include <GL/glext.h>
#elif defined(__APPLE__)
  #include <OpenGL/glext.h>
#endif
  • sparseMatrix.cpp: bug in RemoveRows(int numRemovedRows, int * removedRows, int oneIndexed=0). This issue does not affect the main functionality in Vega, or any of the provided demos. Incorporated into the distribution on Oct 28 2013; prior downloads must manually fix this, by downloading the new version and using the new RemoveRows function.