Unknown dihedral style table

Dear LAMMPS Users,

Hello All. I am using my university cluster to run my LAMMPS simulation. The cluster is using the LAMMPS version, LAMMPS (12 Dec 2018)

In my LAMMPS script, I am tabulating the dihedral potential via

dihedral_style table linear 9999
dihedral_coeff 1 Dihedral_Table_P1_P1_P1_P1.table Read_P1_P1_P1_P1_Dihedral_Table

where my table (entitled Dihedral_Table_P1_P1_P1_P1.table) reads as the following:

N 9999

1 0.0000000000000000e+00 1.4100000000000000e+02 3.2752957995601988e+00
2 1.8001800180018002e-02 1.4094103877948587e+02 7.7009088044470627e+02
3 3.6003600360036005e-02 1.2707801662926613e+02 4.5047428720749252e+02
4 5.4005400540054004e-02 1.1896866852472081e+02 3.1961659323721449e+02
5 7.2007200720072009e-02 1.1321499447904638e+02 2.4791389003169576e+02
6 9.0009000900090008e-02 1.0875209816884484e+02 2.0256039717579677e+02
7 1.0801080108010801e-01 1.0510564637450106e+02 1.7126237555121583e+02
8 1.2601260126012601e-01 1.0202261531147288e+02 1.4835421768599790e+02
9 1.4401440144014402e-01 9.9351972328826648e+01 1.3085769397027960e+02
9999 1.8000000000000000e+02 7.7876485588200435e+00 -1.6326920683199093e-02

I have four columns for my dihedral table as specified in the lammps manual. However, I obtain an error stating: ERROR: Unknown dihedral style table (…/force.cpp:553)

When I use the LAMMPS_64-bit_16Mar2018-MPI (16 Mar 2018) version, it runs without any error.

I looked through the mailing list archives and did not find a thread that discusses the version issue when dealing with dihedral table, and I hardly doubt that looking at the manual, the formatting and the procedure of tabuting dihedral potential has changed.

I was curious if there is a way to resolve this issue.

Thank you.


Masato Koizumi

Hi Masato

dihedral_style table requires you to compile LAMMPS with the USER-MISC
package enabled.

If you did not compile the LAMMPS source code, you will have to learn
how to do that too. (This is a useful skill.) There are 2 ways to do
Method 1: use "make"

Method 2: use "cmake"
(I couldn't find any other source of information for this yet.)


P.S. Also check out:

Dear Mr. Andrew Jewett,

Thank you for your help and guidance.

I apologize for the accident of sending the same message twice. I may have double-clicked the sent button.


Masato Koizumi