The ABINIT Users Mailing List ( CLOSED )

["Matteo Giantomassi" <Matteo.Giantomassi@uclouvain.be>]

> Dear abinit users,
>     I want to calculate Fermi surface using abinit 5.6.3, and I get some
> problems.
> The following message is the problem:
> ===========================================
>  prtfsurf does not work with non-zero kpt shift
>  Action: set nshift 1 and shiftk 0 0 0
> ===========================================
> However, I used nshiftk 1 and shiftk 0 0 0 in my input file.  But when I
> try to
> use some lower version abinit, for example 5.4.4, it runs correctly.
> Is it a bug in 5.6.3?
>
 No, it should be a simple problem with the consistency check done
 in 13iovars/chkinp.F90 probably due to a strict tolerance on the
 values of shiftk. How do you specify shiftk in the input file
 Are you using 0 0 0 or 0.0 0.0 0.0 ?

 Could you add a print in chkinp.F90 in the part of the code
 where the check is done.

  if (sum(abs(shiftk)) > tol8) then
    print*,'DEBUG : ',SUM(ABS(shiftk))  <<< Add this line
    write(message,'(4a)') ch10,&
&    ' prtfsurf does not work with non-zero kpt shift ', ch10, &
&    ' Action: set nshift 1 and shiftk 0 0 0'
    call wrtout(6,message,'COLL')
    call leave_new('COLL')
   end if
  end if

 then recompile, rerun the job and post the result on the forum.
 Thanks in advance

 Best Regards,
 Matteo Giantomassi

>
> Cheng YC
> Nanjing university
>