[Pw_forum] problem regrading convergence

Lorenzo Paulatto Lorenzo.Paulatto at impmc.upmc.fr
Sun Apr 15 19:38:17 CEST 2012


On Sun, 15 Apr 2012 11:50:18 +0200, bramha pandey  
<pandey.bramha at gmail.com> wrote:
> LiInTe2

If this crystal is metallic, at least in DFT and at the specific unit cell  
volume you are using, it will probably not converge without a smearing.

> mixing_beta = 0.8

If you have problems with convergence you should *reduce* not increase  
this parameter (take something like 0.3)

>  diagonalization = 'cg',

this controls the diagonalization performed at each scf step, not the scf  
step iteself, use the default algorithm instead wich is faster.

bests


-- 
Lorenzo Paulatto IdR CNRS @ IMPMC/UPMC - Paris 6
phone: +33 (0)1 44275 084 / skype: paulatz
www:   http://www.impmc.upmc.fr/~paulatto/
mail:  23-24/4é14 Boîte courrier 115, 4 place Jussieu 75252 Paris Cédex 05


More information about the Pw_forum mailing list