[gmx-users] dppc benchmark forcefield
lindahl at stanford.edu
Tue Jan 28 18:25:49 CET 2003
Pedro Alexandre Lapido Loureiro wrote:
>I am using the forcefield attached to dppc1024 benchmark for my own system.
>It is from Berger (Biophys. J. 72,2002-2013,1997) but with two main exceptions:
> LC2 14.02700 0.000 A 0.70067E-02 0.24845E-04
> LC2O 14.02700 0.000 A 0.59496E-02 0.17914E-04
>According to Berger, should be:
> LC2 14.02700 0.000 A 0.85953E-02 0.30478E-04
> LC2O 14.02700 0.000 A 0.72985E-02 0.21975E-04
>Besides, the LJ and Coulomb 1-4 interactions should have been scaled with fudge
>factors 0.25 and 0.5, respectively.
>Forgive me for this (another!) boring question ...
You are right about the fudge factor - we must have missed that when we
updated the benchmark for 3.1.
(The fudge factor used to be given in the mdp file instead); I'll try to
remember to fix it for the next release.
The oxygen parameters are more difficult; I have a vague memory of
checking Oliver (Berger)'s parameters
and finding that the correct OPLS parameters were very slightly
different. It doesn't make a huge difference,
but if it is important I would recommend checking OPLS.
More information about the gromacs.org_gmx-users