The ABINIT Users Mailing List ( CLOSED )

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

Dear Manolo,

In the present state of the code, there is no need to look at the inner 
variables of the subroutines, if you want to perform a fully dynamical 
GW calculation.
Instead, look at the new input variables explained in the "GW input 
variables" file of Abinit version 5.1, in particular:
_  gwcalctyp
_  nfreqim
_  nfreqre
_  freqremax
_  nfreqsp
_  freqspmax

If you want examples of input files, please look at the test files 
numbers 84 - 88 in directory tests/v4/.

In case you have further questions, don't hesitate...

Have nice calculations.


Fabien



mperez@mpi-halle.mpg.de wrote:
> Dear Fabien,
>
> thanks for your email with detailed information and the references (thesis).
> According to the info in the abinit website for version 5.1.2, your
> implementation of GW allows for some "advanced options" not previously
> available. Probably you remember i was asking on extracting the
> quasiparticle states. Well, this is referred to in the website as well as
> the really helpful option to extract the spectral function.
>
>  In order to run fully dynamical (ie expensive) GW, rather than plasmon
> pole model, which is explained in the website on "advanced options"
> (Calculations without using the Plasmon-Pole model ), i seem to remember
> the former version of the code had two inner logical parameters
> analytic_continuation and contour_deformation, which can be used instead of
> plasmon_pole_model, in which case probably no extra input's are required
> (as compared to the tutorials files). But since you developed this new
> version, you probably included extra inputs/options???
> I am just wondering whether for this advanced option (no plasmon pole
> model), i should ask you for some example input file or there is no need
> for that.
>
> thanks
>
> Manolo