I want build a rigid indenter using atom instead of the command of fix indent.
So fix setforce will ouput a 3-vector of total force on the
atoms before you zeroed them. You can verify that
result is correct by dumping forces yourself and summing them.
Note that fix store/force will store forces before you change
them with fix setforce. So those forces are what you would
need to dump or sum via compute reduce.
Assuming LAMMPS is computing it correctly, then why
you don’t get the results you expect at different temps
is a Q about your model, not about LAMMPS.
Steve
Then invoke a compute property/atom to compute the forces and output the vectors in your loop.