[Wien] fourier coefficients in lapw2

[Tran, Fabien]

Hi,


CALL setfft1(n,iff1,iff2,iff3,a(1,je1),fft,keigen)

does not mean that only a(1,je1) is passed. a(1,je1) indicates the position

of the first element of the vector inside setfft1, and the vector is constructed

along the first dimension of a, i.e, a(1:nmat,je1).


F. Tran

________________________________
From: Wien <wien-bounces at zeus.theochem.tuwien.ac.at> on behalf of Marcus Ekholm <marcus.ekholm at liu.se>
Sent: Wednesday, October 30, 2019 2:15 PM
To: Wien at zeus.theochem.tuwien.ac.at
Subject: [Wien] fourier coefficients in lapw2

Hi!

I have a question on how lapw2 calculates the charge density in the interstitial region.

Inside fourir.F (line 294 in version 19.1) we call setfft1 as:

CALL setfft1(n,iff1,iff2,iff3,a(1,je1),fft,keigen)

where a(1,je1) is a single number.
But the subroutine setfft1 expects A to be an array of a different size:

!_REAL      REAL*8           A(nmat)
!_COMPLEX      COMPLEX*16           A(nmat)

and inside setfft1 we then loop over (some of) the elements in A, from 1 to n, to set the fft array.

Where do the elements A(2:n) in settfft1 come from? And in fourir.F why is only a(1, ...) passed to setfft1, why not a(2,...), etc ?


Best regards,
Marcus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20191030/2a7db4a0/attachment.html>