Is it a bug in the code?

I have two identical input files except one has more significant digits than the other in ‘region’ commands to compute surface energy of {111} surface in Al. They report different energies and one is correct. I used version ‘lammps-16Sep12’ on linux machine. Does anybody know what is wrong with the input files?

INPUT FILE I:

I ran your input scripts and surface atoms appear to be missing in the second input script.

Although it doesn’t look that way looking at the dump file, I think you have atoms positioned exactly on your cutting plane. Thus it could come down to significant digits to decide which atoms are created.

Adding
origin 0.0001 0.0001 0.0001
to the second two lattice commands in input file #2 gives the same result for surface energy as input #1.

-Zach