[gmx-users] f(x) g(x) h(x)in the user defined potential functions

[Mark Abraham]

He, Yang wrote:
> Hi Mark,
> 
> Thank you for your reply. From your point of view,   I wonder whether you mean I need to write a table based on different interaction types and then change the source code  for recognition when running my case in the gromacs.

Yes that's what I said.

> I just have no idea about that .Can you give me further suggestion about that? I am stuck by this point

I'm afraid you'll have to look carefully at the data structures GROMACS 
builds during neighbour-searching and build in some hook to allow you to 
use the correct tables later in src/gmxlib/nonbonded.c. You're own your 
own there.

Mark