Re: NPT_NPgammaT_or_NPAT_for_membrane_with_CHARMM_forcefield

From: Jerome Henin (jhenin_at_cmm.chem.upenn.edu)
Date: Sun Nov 11 2007 - 14:40:16 CST

> Please read a basic text book on molecular simulation and statistical
> mechanics. "never simulate using NVT" means, never use NVT for sampling.

The same textbook will tell you about the equivalence of ensembles and
about the fact that thermostats will affect diffusion constants and
other dynamical quantities.

Tutorials give people cookbook-style recipes for solving particular
problems, not one-size-fits-all philosophical principles. They can be
dangerous tools. I would like to insist on the fact that a tutorial is
not a course.

> You need to understand the difference between equilibrium sampling and
> equilibration.

What people called sampling in 10-year-old lipid simulations would not
even be considered equilibration by today's standards. The same may
well apply to our current best efforts, a decade or so from now.

Jerome

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:31 CST