I guess there are two questions:
1) How to create LAMMPS input files (input-script, data-file) so
that you can run a LAMMPS simulation.
As Axel pointed out, the examples/ellipse subdirectory contains a file
"in.ellipse.gayberne" (see attachment) which can be used to run a
LAMMPS simulation (if you have compiled LAMMPS with the "asphere"
package enabled.)
One way to place atoms in a LAMMPS simulation is to use the
"lattice" and "region" and "create_atoms" commands. That is the
method used in the "ellipse" example ("in.ellipse.gayberne" file).
These commands work great if you want to place particles (ellipsoids)
uniformly at points on a lattice. However if instead you want to
manually control the placement (and orientation) of all of the
ellipsoids, then you need to create a "data" file. Axel was saying
that one way to do this is to add the "write_data" command somewhere
before the "run" command, (and optionally comment out the run
commands, since you are only trying to create a data file). The
advantage of this approach (over writing one from scratch) is that it
will create a working data file which you know LAMMPS will accept.
Then you can customize this file.
2) How to modify a LAMMPS input script so that you can view the
simulation trajectory results in OVITO:
To do that, see:
http://www.ovito.org/manual/howto.aspherical_particles.html
I attached the original "in.ellipse.gayberne" and a modified version
which includes these two changes. (But I have not tested the modified
version.)
Andrew
P.S. Note: if you are using the simple, commonly used force-fields
supported by the LAMMPS developers, it is probably okay if you want to
put force-field parameters in the data file (which is what
"write_data" does by default). Otherwise, you should leave the
"coeff" commands in the input script (by using "write_data" with the
"nocoeff" option).
in.ellipse.gayberne (1.54 KB)
in.ellipse.gayberne.modified (2.44 KB)