[Pw_forum] Vc-relax

[Lorenzo Paulatto]

On Fri, 06 Feb 2009 05:31:14 +0100, Lex Kemper <lex at phys.ufl.edu> wrote:

> First, is there a need for a high energy cutoff, or high number of  
> kpoints?

Dear Lex,
your calculation must be well converger wrt kpoints and cutoff; if it is a small system, it is often better to increase cutoff an kpoints a bit more than the bare minimum and don't worry about it anymore.

> Second, having tried using it previously, I've found that the lattice
> constants are updated by very very small increments. What sets the scale
> on the size changes?

It depends on the algorithm you are using. 

-- 
Lorenzo Paulatto
SISSA  &  DEMOCRITOS (Trieste)
phone: +39 040 3787 511
skype: paulatz
www:   http://people.sissa.it/~paulatto/

     *** save italian brains ***
 http://saveitalianbrains.wordpress.com/