[Wien] Large no of K-points in the IBZ

[Peter Blaha]

>  Please, I need to use large number of k-points to do
>  some calculations. When I increase the number of
>  k-points in the whole cell beyound about 40,000, it
> shows "nmax too large in ord1"  without giving the
> k-points in the IBZ.
>
> Please are there parameters to redimension?

We had such questions several times before in the mailing list.

Read the UG (of course the KGEN section):

6.5.2 Dimensioning parameters
The following parameters are used in main.f, ord1.f (static arrays):
 IDKP     number of inequivalent k-points (like NKPT in other programs)
 NWX      internal parameter, must be increased for very large k-meshes
 INDEXM   internal parameter, must be increased for very large k-meshes

cd $WIENROOT/SRC_kgen

grep -i parameter *.f

divisi.f:        parameter (nprim=16)
divisi.f:        parameter (niter=10)
main.f:! AND OFF BY THE PARAMETER "ICOR" IN THE SUBROUTINE "WEIGHTS"
main.f:! SETTING OF THE PARAMETER "TOLMAX" IN SUBROUTINE DOS.
main.f:! COMMENTED OUT IF THE PARAMETER "ICRAY" IN SUBROUTINE "ORD1" IS 0.
main.f:      PARAMETER (NSYMX=50,IDKP=5000,NBX=2,NWX=6600000)
ord1.f:      parameter (indexm=1200000)

Aparently you must look into main.f and ord1.f

edit ord1.f:
        if(nmax.gt.indexm) then
        write(*,*) 'nmax too large in ord1',nmax
        stop 'nmax'
        end if

The program should have told you, how large   "indexm" should be.

Set it larger and recompile.
Eventually you will need to increase also other parameters (IDKP, NWX),
the program will "tell" you.


                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------