[Wien] question about IDKP settings and Lapw2 error

[Peter Blaha]

You have to be systematic.
Do an optic calculation for 10000 k, plot the results.
Increase k-points to 50000 k, plot the results and COMPARE to previous 
run. If there is a substantial change,
increase k-points again by a factor of 5, ....

If you go in such steps, you also will hit the "redimension problem", 
but you can probably master it.

---------------
 >   'LAPW2' -        filename: /fccni7.vectorup

This looks as if you have defined:
SCRATCH=/    instead of
SCRATCH=./

On 09/09/2014 09:51 AM, bruce.tian wrote:
> Dear all:
>
> I am using large kpoints (5000000, similar to the number of literature
> by Claudia) to calculate optic properties of fcc Ni with SO included.
> However I have two problem in running the program:
>
> 1. After the k-point parallel calculation of SCF. I changed the TETRA
> value to 101 in case.in2. Then run 'x kgen' and get the following error
> message:
>
> x kgen
>    NUMBER OF K-POINTS IN WHOLE CELL: (0 allows to specify 3 divisions of G)
> 5000000
>   length of reciprocal lattice vectors:   1.653   1.653   1.653 170.998
> 170.998 170.998
>    Shift of k-mesh allowed. Do you want to shift: (0=no, 1=shift)
> 1
>   IDKP in main.f too small.
>    Set it to expected # of k-points in IBZ (gt.       60001 )
> reduz.f: IDKP in main.f too small
> 21.361u 43.618s 1:12.48 89.6%   0+0k 0+502904io 105pf+0w
>
> But each time I increase the IDKP in main.f according to the number
> shown in the error message and run x kgen the program will ask for
> another larger number. Shall I increase the IDKP dramatically much
> larger than the number required in prompt or change some other settings?
>
> 2.After the k-point parallel calculation of SCF and change the TETRA
> value to 101 in case.in2 I run "x lapw -fermi -up" with SO selected by
> w2web and I get the erro message as following:
>
> Error in LAPW2
>   'LAPW2' - can't open unit: 10
>   'LAPW2' -        filename: /fccni7.vectorup
>   'LAPW2' -          status: unknown      form: unformatted
>
> Furthermore I found all x lapw2(1) can not work. What is the problem?
> For the large k-points case, can x lapw2, x optic, x joint be run in
> k-points parallel mode by w2web?
>
> Thanks!
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WWW: 
http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------