update documentation for vashishta pair style

This commit is contained in:
Axel Kohlmeyer 2016-09-08 17:20:52 -04:00
parent fe581e8ced
commit 71eed1d612
1 changed files with 9 additions and 2 deletions

View File

@ -9,15 +9,20 @@
pair_style vashishta command :h3
pair_style vashishta/omp command :h3
pair_style vashishta/table command :h3
pair_style vashishta/table/omp command :h3
[Syntax:]
pair_style style args :pre
style = {vashishta} or {vashishta/table}
style = {vashishta} or {vashishta/table} or {vashishta/omp} or {vashishta/table/omp}
args = list of arguments for a particular style :ul
{vashishta} args = none
{vashishta/omp} args = none
{vashishta/table} args = Ntable cutinner
Ntable = # of tabulation points
cutinner = tablulate from cutinner to cutoff
{vashishta/table/omp} args = Ntable cutinner
Ntable = # of tabulation points
cutinner = tablulate from cutinner to cutoff :pre
@ -60,7 +65,9 @@ equally spaced in R^2 space from cutinner^2 to cutoff^2. For the
two-body term in the above equation, a linear interpolation for each
pairwise distance between adjacent points in the table. In practice
the tabulated version can run 3-5x faster than the analytic version
with little loss of accuracy for Ntable = ???.
with with moderate to little loss of accuracy for Ntable values
between 10000 and 1000000. It is not recommended to use less than
5000 tabulation points.
Only a single pair_coeff command is used with either style which
specifies a Vashishta potential file with parameters for all needed