[gmx-developers] MC integrator

[Andre Assunção S. T. Ribeiro]

Dear Developers,

I have been working on a Monte Carlo integrator for GROMACS for some
time. I have now finished a working version. It is still very crude,
only treats rigid body translation/rotation and performance could be
much better. I have adapted the code for parallel runs, but only with
particle decomposition. I have tested it using a box with SPC water in
both NVT and NPT ensembles and the results look fine. I would appreciate
if anyone could take a look at the code and help me test it and/or give
any suggestions.

The code can be downloaded from:
git://github.com/aastr/GROMACS.git



The following mdp file has been used for NPT jobs:

title             =  mc
cpp               =  /usr/bin/cpp
coulombtype       =  PME
constraints       =  none
integrator        =  mc
dt                =  0.002
nsteps            =  50000000
nstlist           =  1
ns_type           =  grid
rlist             =  0.9
rcoulomb          =  0.9
rvdw              =  0.9
fourierspacing    =  0.12
fourier_nx        =  0
fourier_ny        =  0
fourier_nz        =  0
pme_order         =  6
ewald_rtol        =  1e-5
optimize_fft      =  yes
nstxout           = 1
nstenergy         = 1

tcoupl                   = berendsen
tc_grps                  = sol
tau_t                    = 0.1
ref_t                    = 300

Pcoupl = MC
ref_p = 1.0
pcoupltype = isotropic
tau_p = 0.5
compressibility = 4.5e-5

; MC stuff
delta_v = 0.05
delta_r = 0.012
delta_phi = 15.0
nst_p = 400


Best Regards,
Andre Ribeiro
UFRJ - Brazil