[MMTK] Fwd: Question about OrthorhombicPeriodicUniverse

Konrad Hinsen research at khinsen.fastmail.net
Sat Feb 9 08:47:14 UTC 2013


mmtk-bounces at starship.python.net writes:

 > I’m running MMTK on a Win32 platform with Python 2.7.3.  I was just
 > trying to build a periodic box of water. I can build the array of
 > water just fine, and run a small dynamic simulation, if I use
 > “InfiniteUniverse”.  Now when I try to run a
 > “OrthorhombicPeriodicUniverse” the energy evaluation seems to go
 > crazy.  An example of the output is

If you get reasonable results in an InfiniteUniverse but NaN in a
periodic universe, then most probably your initial configuration is
not compatible with the periodic boundary conditions. There are
probably clashes between some atoms inside your box and the periodic
images of other atoms. Unfortunately this is not easy to spot by
visualization.

A first step towards analyzing what's going on is

  print universe.energyTerms()

That will show you which of the energy terms become NaN. If it's one
of the nonbonded terms (Lennard-Jones, electrostatic), that supports
my hypothesis. The bonded terms should be exactly the same as they are
in an InfiniteUniverse.

Konrad.
-- 
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: research AT khinsen DOT fastmail DOT net
http://dirac.cnrs-orleans.fr/~hinsen/
---------------------------------------------------------------------



More information about the mmtk mailing list