Page 1 of 1

internal error in subroutine INVGRP

PostPosted: Fri Oct 28, 2005 3:32 am
by Brane
Error message:
VERY BAD NEWS! internal error in subroutine INVGRP:
inverse of rotation matrix was not found (increase SYMPREC) 2

I encountered this problem when I try to do a single point energy calculation after geometry relaxation (use the CONTCAR generated). I think it is not due to the structure I used. The error is still there after increasing SYMPREC. BTW: The structure used is a hexagonal one.

Any comment about this error is appreciated.

internal error in subroutine INVGRP

PostPosted: Fri Nov 25, 2005 2:57 pm
by admin
most probably, your k-mesh breaks the symmetry, this sometimes happens
for even k-meshes for hexagonal or low-symmetry lattices.
1) use an odd k-mesh.
if the error remains,
2) upgrade your VASP version, if necessary
3) decrease the optimisation of symlib.F and symmetry.F to -O0 and recompile