need more than 200 group

dear users

i want to write a multiscale code with matlab and lammps and i must use more than 200 groups of atoms for add force and compute.

i read previous question but i didn’t find the solution

can any one help me and offer me a clear way

Best regards,

kashani

There may be a good solution for you with current lammps capabilities. But without more info, I doubt your going to get much in terms of useful advice.

dear users

i want to write a multiscale code with matlab and lammps and i must use
more than 200 groups of atoms for add force and compute.
i read previous question but i didn't find the solution

​if you have researched the mailing list archive, you should have already
read that it would be a *very* non-trivial operation to extend the number
of groups beyond the current limit. in addition, you should have found the
response that thus far, nobody has come up with a request for more groups
that justifiably could not be handled *better* in a different fashion.

there are multiple simple ways to group atoms, e.g. via molecule ids and
using compute chunk/atom. you can also add equivalent arbitrary per atom
properties that migrate with atoms via fix property/atom. you can do the
grouping within your code etc. etc. LAMMPS is extremely flexible and groups
are a rather clunky way to solve grouping problems, especially for coupling
to another code.

please also note that if you had to process over 200 fixes and computes,
you would have to pay a serious performance penalty.

axel.