Dear everyone,
I have a naive question about the number of k points in PHonon calculations.
I have a magnetic system (nspin=2) under calculation for the phonon DOS. The crystal structure is bcc-like with space group number of 229 (a supercell of bcc lattice, which contains a single impurity atom) . I observe that in the output of ph.x the number of k points is doubled, compared with what was given by pw.x. I also observe that the number of k points remain the same with both codes for non-magnetic system. Does that mean magnetic systems involve some symmetry operations which must be discarded during the DFPT calculation? And does that mean the ratio of the numbers of k points must be 2:1 (N_ph : N_pw) ?
(I'm using 5.4.0, if this matters)
Also another naive question. How should I estimate the time cost of PHonon calculations?
In the guide, it is said that "each of the 3Nat modes requires a time of the same order of magnitude of self-consistent calculation in the same system". From the source code, it appears a nscf calculation usually is necessary. So would the following formulation be correct,
For each q point : t_total = t_nscf + sum_over_representations (Num_modes_each_representation * t_scf)
Many thanks for your explanation
Yi Wang
----------------------------------------------------------
Nanjing University of Science and Technology, China
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
