[lammps-users] a short question about restart file

Dear all,
If I re-specified the pair_style and pair_coeff commands(lj/charmm pair styles are used) in an input script that reads a restart file,
will this make my results wrong?
I remember that in the mailist someone discussed: if the same group name is re-specified after reading a restart file, lammps will give some tricky results,
is this right?

thanks a lot!
xiong

Dear all,
If I re-specified the pair_style and pair_coeff commands(lj/charmm
pair styles are used) in an input script that reads a restart file,
will this make my results wrong?
I remember that in the mailist someone discussed: if the same group
name is re-specified after reading a restart file, lammps will give some
tricky results,

dear xiong,

this is basically impossible to answer without a specific example.

...and it is _extremely_ easy for you to verify what kind of behavior
you would get under the circumstances that you are referring to.

all you have to do is to create a very minimal input (a handful of atoms?)
and create multiple input variants to systematically test whether adding
a statement makes a change or not. with that small an input, you can
monitor everything for every step and the calculation will be very fast.
just see it as a scientific experiment.

cheers,
   axel.

The restart file sets the pair style (for most pair styles, some
that read files do not, see the doc page for the pair style).

If you re-use the pair_style and/or pair_coeff commands in
your input script after reading the restart file, then you are
setting the pair style to something new. Same as if you
used the pair_style command twice in a script that didn't
read a restart file. Ditto for groups. If you use the group
command multiple times in your input script, the group
command does what you specify.

There's nothing complicated here. Think of replacing the read_restart
command with all the commands that were in the other
input script that led up to the writing of the restart file.

Steve

Steve