[Pw_forum] Intermediate images in NEB

[Axel Kohlmeyer]

On Mon, 1 Oct 2007, Daniele Passerone wrote:

dear daniele,

please complete your bug report by stating which
version of quantum espresso you are using, and
if you have the time, you could also check whether
the bug still exists in the cvs version...

ciao,
  axel.

DP> Dear forum members
DP> 
DP> We noticed a small bug in the neb capabilities of pwscf, or maybe we
DP> are missing something.
DP> Indeed, the program seems to set always a linear interpolation between
DP> first a last image, even when the
DP> intermediate images are given.
DP> 
DP> The origin seems to be in the subroutine initial_guess (module
DP> path_base.f90)
DP> where an array pos_n with the linearly interpolated position is built.
DP> 
DP> The purpose should be to copy pos_n(:,:) in the pos(:,:) vector only if
DP> the intermediate images given by the user are absent or 
DP> unreasonable.
DP> Unfortunately,
DP> there is a line
DP> 
DP> pos(:,:) = pos_n(:,:)
DP> 
DP> that always copies the linear guess into the position array.
DP> 
DP> For the moment i modified this line into 
DP> 
DP>         if (input_images .ne. num_of_images) then
DP>              pos(:,:) = pos_n(:,:)
DP>          endif
DP> 
DP> And that should do the job.
DP> 
DP> 
DP> Best regards,
DP> 
DP> Daniele Passerone
DP>  
DP> _______________________________________________
DP> Pw_forum mailing list
DP> Pw_forum at pwscf.org
DP> http://www.democritos.it/mailman/listinfo/pw_forum
DP> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.