[gmx-users] Coulomb interactions

[Mark Abraham]

On 15/03/11, Swarnendu Tripathi  <stripat1 at kent.edu> wrote:
> Hello everyone,
> 
> In gromacs when the Coulomb interactions are calculated are they calculated for the all atom pairs or only for the charged atoms pairs? 
> 

GROMACS notices atoms with zero charge and does not put them into lists scheduled for evaluating charge-charge interactions. Similarly for zero LJ parameters.


> If the Coulomb interactions are not calculated only for the charged atoms then how can I modify the gromacs to do that. In my system I have only 27 charged atoms  and I want gromacs to calculate  the Coulomb interactions only for these atoms. I am using gromacs 4.0.7.
> 

Obviously, this is only an efficiency issue, and GROMACS already takes care of it. You could see that this is true by making all of your "uncharged" atoms have charge 0.000001 and observing the increase in run time. You can also see this from the breakdown of computation time at the end of your .log file. There will be significant time spent doing LJ-only loops, and less time doing LJ+Coulomb loops.

Mark



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110315/9a1ed8f4/attachment.html>