104
Styles with a {cuda}, {gpu}, {omp}, or {opt} suffix are functionally
105
the same as the corresponding style without the suffix. They have
106
been optimized to run faster, depending on your available hardware, as
107
discussed in "Section_accelerate"_Section_accelerate.html of the
108
manual. The accelerated styles take the same arguments and should
109
produce the same results, except for round-off and precision issues.
104
Styles with a {cuda}, {gpu}, {intel}, {kk}, {omp}, or {opt} suffix are
105
functionally the same as the corresponding style without the suffix.
106
They have been optimized to run faster, depending on your available
107
hardware, as discussed in "Section_accelerate"_Section_accelerate.html
108
of the manual. The accelerated styles take the same arguments and
109
should produce the same results, except for round-off and precision
111
These accelerated styles are part of the USER-CUDA, GPU, USER-OMP and OPT
112
packages, respectively. They are only enabled if LAMMPS was built with
113
those packages. See the "Making LAMMPS"_Section_start.html#start_3
114
section for more info.
112
These accelerated styles are part of the USER-CUDA, GPU, USER-INTEL,
113
KOKKOS, USER-OMP and OPT packages, respectively. They are only
114
enabled if LAMMPS was built with those packages. See the "Making
115
LAMMPS"_Section_start.html#start_3 section for more info.
116
117
You can specify the accelerated styles explicitly in your input script
117
118
by including their suffix, or you can use the "-suffix command-line