Axel,
stefan,
Thank you for all the help.
I started writing the code, but I can not find any documentation about
ev_tally*. Is there any documentation about it besides the code
itself?
only comments in the code. the purpose of the ev_tally*() family
of methods is to "tally" (=accumulate/collect) global or per atom
properties like the potential energy or the virial into the
proper arrays (or not, depending on different flags).
And a second question. I am going to add some code to bond_fene that
comes from reading pair_dipole_cut. I have calculated all the
energies, forces and torques that my system is going to have
(according go my model) and I am using dipoles (hybrid angle dipole).
Is there anything else that I have to do besides not screwing up
ev_tally*. I'm sorry about repeating the question, but I don't
understand the meaning of "propagator" (English is a second language
for me and I have not found the word in a technical dictionary).
http://en.wiktionary.org/wiki/propagate
in this context propagator is the algorithm that moves "stuff" forward,
e.g. the velocity verlet. the issue is that you can easily come up with
a model and energies/forces, but unless your particles have moments of
inertia and you use a propagator that can rotate them around their
center of mass, there is very little use to it.
And lastly. I'm still a student and this is my first assignment in MD.
As I am modeling something fairly simple and lammps is quite powerful,
while you think of it as simple, the solution you propose is
fairly non-standard and not that commonly used. as a beginner,
it would be much easier to first understand the basics of MD
with a much simpler systems. the classic example would be
liquid argon with lj particles. this is a very well studied
system and you would first learn how to control all kinds of
simulation parameters to reproduce well known results, before
you apply that experience to your research project. if you
start with the research right away, one problem will continuously
be: is what you see correct? and if not, is it wrong because of
errors in the programming, or errors in the way how the simulation
is run.
it seems to me that I should not need to write supplemental code. Is
there any obvious mistake in my method (modeling torsion resisting
monomers in this manner)? I am asking for an opinion here, as this is
the first MD code I am studying.
i repeat myself, but in my opinion you have not yet fully understood
the implications of the difference between point particles and
extended particles and difficulties resulting from propagating the
latter. even though you often visualize atoms as spheres looking like
billiard balls, they are not acting like that in their physics.
cheers,
axel.
Where else should I