On Jul 30, 2009, at 18:47 , Cristian Degli Esposti Boschi wrote: > using the phonon code I noticed that, even > for simple molecular structures, the dynamical > matrix computed by ph.x is not symmetric [...] > In general, this does not guarantee real > eigenvalues. How are the eigenfrequencies > computed then?
the dynamical matrix is symmetrized before diagonalization. There is a check that the unsymmetric part (coming from inexact self-consistency) is small. If not, a message is issued Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
