The ABINIT Users Mailing List ( CLOSED )

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

Hi,

Usually Abinit understand input structures correctly. If it tells you
that your input is monoclinic, it is quite likely this is true and it
may matter a lot for your calculation if it is supposed to be
orthogonal. Sometimes this kind of errors results from lack of
numerical precision in the input file. Don't forget to use fraction,
sqrt ... wherever relevant . The best for other people to tell you
what is going wrong is to post your input file at least. May be some
part of your output would be usefull too.

About GW calculation: For as much as I know you just have to give
Abinit the coordinates of each k-point you want to include. For
instance if you want to use standard monkhorst pack gris plus some
k-point you can first generate the list of coordinates for the
monkhorst pack and then use kptopt 0, nkpt and kpt to specify the
point you are interested in.

regards

PMA

On Sun, Mar 23, 2008 at 11:22 AM, Sophie Green
<us_sophie.green@yahoo.com> wrote:
> I am a new user to Abinit. I looked through the tutorial. I want to
>  calculate band gap with GW. I gave the conventional unit cell parameters
>  (acell) and rprim, also the atomic positions. But the output file shows
>  that the space group is monoclinic (P2_1/c). But it should be
>  orthogonal (cmca). I don't know why the space group is wrong. Will this
> cause
>  mistake in the following calculations?
> Besides, If the point I want to calculate in GW is nonspecific point,
>  what should I do to conclude this point in KSS calculation?
> If anyone has this experience, please instruct me. Thanks in advance.
>
>
>  ________________________________
>
> Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it
> now.



-- 
Pierre-Matthieu Anglade