[QE-users] Keep the symmetry during a calculation

[Lorenzo Paulatto]

Actually, everything is in my gitlab fork.

You need to take this version of Modules/latgen.f90:
https://gitlab.com/paulatz/q-e/raw/develop/Modules/latgen.f90
which contains an additional subroutines to recompute the new values of 
celldm and regenerate the unit cell (which can change it slightly), than 
call this new subroutine just before recips/volume in 
PW/src/move_cell.f90 and PW/src/vcsmd.f90
like I did here:
https://gitlab.com/paulatz/q-e/commit/2458751fa28c0af7f01c3dfcef220ef733b89b82

You can just get my version of move_ions.f90 here:
   https://gitlab.com/paulatz/q-e/raw/develop/PW/src/move_ions.f90
but I'm not sure it works with anything else than the git version of QE.


On 11/25/18 2:51 PM, Isaiah Moses wrote:
> Yes please, Dr Lorenzo.
> I've been struggling for weeks with keeping symmetry of structures in 
> vc-relax.
> I'll appreciate your "tiny patch".
> Isaiah
> 
> On Sun, Nov 25, 2018, 4:42 AM Lorenzo Paulatto <paulatz at gmail.com 
> <mailto:paulatz at gmail.com> wrote:
> 
>      > 2) If so, is there a way to do what I want? Keep the symmetry,
>     and get a
>      > final structure that i still in the cubic form?
>      >
> 
> 
>     If you are still interested, I have written a tiny patch that conserves
>     the "ibrav representation" of the crystal during the vc-relax
>     procedure,
>     it takes a few more iterations to optimize the structure, but it seems
>     to be stable. I'd like some feedback if anybody wants to test it.
> 
>     cheers
> 
> 
> 
>     -- 
>     Lorenzo Paulatto - IMPMC - CNRS/SU - Paris
>     _______________________________________________
>     users mailing list
>     users at lists.quantum-espresso.org
>     <mailto:users at lists.quantum-espresso.org>
>     https://lists.quantum-espresso.org/mailman/listinfo/users
> 
> 
> _______________________________________________
> users mailing list
> users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
> 

-- 
Lorenzo Paulatto - Paris