Hi
I am coding a new pair style and in the course of testing the model we discovered that we can’t run ewald sums with our pair style. What do I need to do to the code to make it compatible?
thanks much
matthew
Hi
I am coding a new pair style and in the course of testing the model we discovered that we can’t run ewald sums with our pair style. What do I need to do to the code to make it compatible?
thanks much
matthew
Hi
I am coding a new pair style and in the course of testing the model we
discovered that we can't run ewald sums with our pair style. What do I need
to do to the code to make it compatible?
well, first of all, you need to code the corresponding real-space part
of the coulomb interactions, see lj/cut/coul/long or lj/long/coul/long
or lj/long or lj/cut/coul/msm or lj/cut/tip4p/long and so on.
also, there are a bunch of flags defined in pair.h, that you can set
in the constructor of your pair style to signal, which kinds of
long-range interactions, your pair style is compatible with:
int ewaldflag; // 1 if compatible with Ewald solver
int pppmflag; // 1 if compatible with PPPM solver
int msmflag; // 1 if compatible with MSM solver
int dispersionflag; // 1 if compatible with LJ/dispersion solver
int tip4pflag; // 1 if compatible with TIP4P solver
int dipoleflag; // 1 if compatible with dipole solver
int reinitflag; // 1 if compatible with fix adapt and alike
again, please see the corresponding pair style implementations for examples.
axel.