[QE-users] The grid for wave functions in real space;

Paolo Giannozzi p.giannozzi at gmail.com
Wed Jan 9 10:56:25 CET 2019


On Fri, Jan 4, 2019 at 6:51 PM Taghipour, Zahra <taghipour.5 at osu.edu> wrote:

After reading the developer’s manual and the conversations in the forum, I
> realize that the grid in which QE uses to express the wave functions is the
> “smooth grid” determined by nr1, nr2, nr3 (right?)
>

almost: nr1,nr2,nr3 (or dfftp%nr1,2,3) are the dimensions for the "dense"
grid (charge density including augmentation charges and potential); the
dimensions for the "smooth" grid (for \psi, V\psi and |\psi|^2) are
nr1s,nr2s,nr3s (or dffts%nr1,2,3). Note that:
- the physical dimensions of the FFT arrays contain a final x (e.g.,
nr1x,nr2x,nr3x) and may (but usually don't) differ from the true
dimensions
- in parallel execution, FFT arrays are distributed across processors (in
real space, the last dimension is).
You may take inspiration from the calculation of the dipole in routine
Modules/compute_dipole.f90 (ignore the "minimum image convention": r is not
periodic so the dipole is well defined only for finite systems)

Paolo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20190109/bd841063/attachment.html>


More information about the users mailing list