The ABINIT Users Mailing List ( CLOSED )

["Anglade Pierre-Matthieu" <anglade@gmail.com>]

OK,

- What is the evolution of your energy with respect to acell?
- Are you sure that you convergence with respect to the k-point mesh
is fine enough?
I mean do the errors you get when you change acell are bigger or
smaller than the absolute convergency of your k-point mesh?
- I don't know the cell parameter of Ag and the structure of your
cluster. Yet I guess that there is less than 10 bohr between each
cluster when your acell is 20 bohr. I'd guess that this is  too small
to some extent.
-Last question: Do your calculation reach convergency in 200 step?
- I've seen that ou've increased nline. That's nice but I think that 
it would be more efficient, in order to reach a real convergency, to
increase nnsclo also.

--
Pierre-Matthieu Anglade