more than 32 groups

Dear Sir,

I want use fix move to give some random moves for each molecules.
In order to do so, I need to assign each molecule with a group ID.
But the question is that I cannot give more than 32 groups.
Could I modify the source code to break this limitation?
And some other suggestions to accomplish the goal are really appreciated.
Thank you very much.

Dear Sir,

   I want use fix move to give some random moves for each molecules.
   In order to do so, I need to assign each molecule with a group ID.

no, you don't.

   But the question is that I cannot give more than 32 groups.
   Could I modify the source code to break this limitation?

no. its non-trivial to do and it is not necessary.

   And some other suggestions to accomplish the goal are really
appreciated.

have a look through the mailing list archives. this is a
regularly asked question and there are several suggestions
on how to deal with it that were given in the past.
so far practically every request for having more groups
turned out to be ill conceived and better done differently.

cheers,
     axel.

Dear Axel.

Many thanks for your reply.
But I want to give different movement for each molecule by fix move.
In fix move command, one can give a specific movement only for a single group.
Here, each group means a single molecule.
So is there any other choice ?
Thanks again.

*have* you searched through the mailing list archives?

axel.

Dear Axel

I do have.
But I still have no idea how to do that.

Even if the group limit was extended to be 1000
(which would be hard and not that useful),
using fix move to do this is a bad idea.

That is b/c the way fix move works is it loops
over all the atoms and does an operation on
those in the group. If the group limit was 1000,
users would think they could use put
1000 fix move commands in their script, one
per group. Which would mean every timestep,
all the atoms would be looped over 1000 times.

You would be much better off to write your own fix,
using fix move as a starting point, that looped
over the atoms once, and did whatever you decide
to each atom based on the molecule ID it had,
or something like that.

Steve