[gmx-developers] [Fwd: Warnings/ Selection of Nonbonded potential.]

Mehmet Suezen suzen at theochem.tu-muenchen.de
Wed Feb 27 15:30:25 CET 2002


David van der Spoel wrote:
> I think you will have to edit grompp too... The check you found should be
> removed, but then it is important to have the Buckingham and LJ separated,
> so before you go into convparm to convert the parameters to a different
> format you probably have to rearrange them such that the default type

I do have changes in grommp as :
in function  renum_atype; 
check for plist

 /* Renumber nlist */
  if (plist[F_LJ].nr > 0)
    ftype=F_LJ;
   else if (plist[F_BHAM].nr > 0)
    ftype=F_BHAM;
   else
    ftype=F_NEWPOT;


But It was a guess. Other relevant function is 
convert_params has also modified as;

  enter_function(&(nbtypes[F_NEWPOT),(t_functype)F_NEWPOT,idef,
                 &maxtypes,TRUE,TRUE);

What did you meant by "rearrange parameters"? It was not clear to me.
And I guess, I'm missing where is those topology file nonbonded
potential
actuall parse in?

> comes first and the other after that. To keep it somehow comprehensible, I
> suggest you work towards having a full Bham matrix and a full LJ matrix.

I assume it is non bonded parameter matrix. In grommp it is in function 
check_pairs. Is it correct?

Many thanks for the comments

Mehmet



More information about the gromacs.org_gmx-developers mailing list