Hi lammps users
I am using rerun command for psuedo simulation. I have finished the real one before I use rerun command, and I also define fix nve for a group of atoms in the real simulation. During the psuedo simulation, I did not unfix the previously defined nve, because “Fixes that perform time integration, such as fix nve or fix npt are not invoked, since no time integration is performed” according to the manual. However, when I use info command to output the fix information during the psuedo simulation, the fix nve is still there.
Should I use unfix command for previously defined nve when I perform the psuedo simulation with rerun?
Fan Li
Hi lammps users
I am using rerun command for psuedo simulation. I have finished the real
one before I use rerun command, and I also define fix nve for a group of
atoms in the real simulation. During the psuedo simulation, I did not unfix
the previously defined nve, because "Fixes that perform time integration,
such as fix nve or fix npt are not invoked, since no time integration is performed"
according to the manual. However, when I use info command to output the fix
information during the psuedo simulation, the fix nve is still there.
yes. this is all consistent with the documentation. having a fix defined
and having it invoked are two completely different things.
Should I use unfix command for previously defined nve when I perform the
psuedo simulation with rerun?
well, do you see a difference with the fix still defined or unfixed?
axel.
I compared the results with fix still defined and fix undefined. No difference.
So, it means that also the fix is still there, but lammps does not invoke it during the psuedo simulation. Is that right?
I compared the results with fix still defined and fix undefined. No
difference.
So, it means that also the fix is still there, but lammps does not invoke
it during the psuedo simulation. Is that right?
what does the manual say?
axel.