On Tue, July 14, 2009 08:29, dev sharma wrote: > occupations='smearing', smearing = > "methfessel-paxton",degauss=0.05,
Dear Dev, your value of degauss is quite high, and (if I remember correctly) the M-P recipe for the smearing can produce some small negative occupancies, which the diagonalizing algorithm does not like. Try to reduce the smearing (which is in Ry units!), or use gaussian smearing; I would suggest the first solution, as it is more likely to produce meaningful results. cheers P.S. please provide your full affiliation when writing to this mailing list -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/
