On 09/28/2013 02:44 PM, Paolo Giannozzi wrote: > in principle, non-scf is faster. In practice, there isn't a > large difference. It seems to me very strange that a non-scf > calculation performed under exactly the same conditions is > much slower than the scf one
I had the impression that SCF would diagonalize the empty bands with reduced accuracy, while NSCF would diagonalize all the bands with full accuracy. However, quickly checking the source code I found no trace of my memory; possibly I did not recall correctly. best regards -- Dr. Lorenzo Paulatto IdR @ IMPMC -- CNRS & Universit? Paris 6 phone:+33 (0)1 44275 084 / skype: paulatz www: http://www-int.impmc.upmc.fr/~paulatto/ mail: 23-24/4?16 Bo?te courrier 115, 4 place Jussieu 75252 Paris C?dex 5
