about the command "fix rigid/nve/small molecule"

here is the introduce:
For bodystyle molecule, each set of atoms in the fix group with a different molecule ID is treated as a rigid body.

if one atom is not glued to the other, i mean ,maybe far away from each other,i assign different molecule id for the two atoms .Do the command treat the two atoms as a rigid body?

here is the introduce:
For bodystyle molecule, each set of atoms in the fix group with a different
molecule ID is treated as a rigid body.

if one atom is not glued to the other, i mean ,maybe far away from each
other,i assign different molecule id for the two atoms .Do the command treat
the two atoms as a rigid body?

no.

here is the introduce:
For bodystyle molecule, each set of atoms in the fix group with a different
molecule ID is treated as a rigid body.

perhaps this sentence is more easily understood if reformulated like this:

For bodystyle {molecule}, each set of atoms in the fix group with the
same molecule ID is treated as a different rigid body.

or with a little more detail:

For bodystyle {molecule} atoms are grouped into rigid bodies by their
molecule ID, i.e. each set of atoms in the fix group with the same
molecule ID is treated as a different rigid body.

2CF6FE5E@...6983....pnglike this,all of the atoms are attach to each other,the number is molecule id .Will the command treat these atoms as eight rigid bodies?3708FA0A@...6982....pngjust like this?

like this,all of the atoms are attach to each other,the number is molecule
id .Will the command treat these atoms as eight rigid bodies?just like
this?

​no. you have two molecule IDs, so there will be two rigid objects.

axel.​

3708FA0A@...6982....png

2CF6FE5E@...6983....png

Thanks!I thought the command treat atoms as a rigid body with different molecule id at first!

3708FA0A@...6982....png

2CF6FE5E@...6983....png

Thanks!I thought the command treat atoms as a rigid body with different
molecule id at first!

it said *sets* of atoms. also, it doesn't make sense to combine different
molecule ids into one rigid objects: how would LAMMPS be able to tell,
which goes with which?
i will propose to have my reformulated expression​ used in the manual to
reduce confusion.

axel.

3708FA0A@...6982....png

2CF6FE5E@...6983....png