The ABINIT Users Mailing List ( CLOSED )

[Fabien Bruneval <fabien.bruneval@cea.fr>]

I recently observed this type of behavior for a finite system.
I concluded that the extrapolar method was too strong for the large 
G-vectors.
Maybe in the future, one could think of an extrapolar parameter that 
depends on G...

By now, I advise you to increase the parameter gwencomp in order to 
reduce the magnitude of the correction.
(Maybe you'll have to increase slightly the nband as a consequence).

Let us know...

Fabien



Manuel Pérez Jigato wrote:
> Dear all
>
> i would like to ask whether the use of the extrapolar method is known 
> to have an important effect on the convergence properties of other 
> paremeters.
> A large increase in ecuteps (the size of the dielectric matrix) seems 
> to accompany the extrapolar method (as compared to the full-contiuum 
> calculation).
> I mean, about twice as much (this implying a memory increase).
>
> Is there any experience with that?
>
> thanks in advance