[lammps-users] Extending lammps

LAMMPS LAMMPS Mailing List Mirror
1

Hello all,

I have been modifying various sections of lammps to suit my need. However I am working on my biggest change thus far, and would like advice with regards to the definition of new atom types. I don't understand what

  comm_x_only
  comm_f_only
  size_forward
  size_reverse
  size_border
  size_velocity
  size_data_atom
  size_data_vel
  xcol_data

are and how to compute what they are, for most of the other sections I could work out what is needed from examples but this bit is not very clear (to me at least).

Any help would be appreciated,

Thank you,

Rizgar Mella

2

Hello all,

I have been modifying various sections of lammps to suit my need. However I am working on my biggest change thus far, and would like advice with regards to the definition of new atom types. I don't understand what

can you please elaborate a little more, what you mean with
"new atom type". lammps is already very flexible and supports
a lot of different calculations, so there is a good chance that
there is already support for what you are looking for.

also, how about making your other modifications available
to the rest of the lammps users?

comm_x_only
comm_f_only
size_forward
size_reverse
size_border
size_velocity
size_data_atom
size_data_vel
xcol_data

are and how to compute what they are, for most of the other sections I could work out what is needed from examples but this bit is not very clear (to me at least).

those are methods to do communication of particle data between
the different domains and the respective sizes of the data blocks
to be communicated. should be fairly obvious from the names and
right in your face obvious from looking at the code. :wink:

axel.

3

See atom_vec.h which documents all the
variables you list.

Steve