Dear lammps developers,
I was wondering if one more quantity could be added to this compute later: the number n_k of face with k edges. The quanty (n_3,n_4,n_5,n_6), i.e., the number of faces with 3, 4 , 5 and 6 edges, is usually reported in literature.
Also, I guess the description of the 2nd quantity in the manual might be a bit confusing. The manual says:“Two quantites per atom are calculated by this compute… The second is the number of faces of the Voronoi cell, which is also the number of nearest neighbors of the atom in the middle of the cell.” Since Voronoi cell is similar to Wigner-Seitz cell which contains only one atom, the faces of Voronoi cell will just be the number of nearest neighbors, but not in the middle of the cell.
Thanks.
Chunguang
Please ignore the comment on the manual description. Thanks.
a) If I understand you, that’s not one quantity, it’s one for each k from 3 to some large number?
i.e. why stop at 6? that doesn’t seem very convenient to include in the output
b) Daniel can comment on whether that info is readily available from the
output of Voro++.
Steve
a) If I understand you, that's not one quantity, it's one for each k from 3
to some large number?
i.e. why stop at 6? that doesn't seem very convenient to include in the
output
Would this be needed on a per atom basis? The data is available, and I
could imagine adding it as a global sum. For per atom this would be a
lot of data and I cannot imagine an advantage of having this data at
runtime. I would suggest obtaining these statistics in postprocessing.
b) Daniel can comment on whether that info is readily available from the
output of Voro++.
See above. The data is available.
Daniel