Re: Can NAMD be used as for general Newtonian mechanics simulations?

From: Vincent Kraeutler (vincent_at_kraeutler.net)
Date: Sat Jul 28 2007 - 07:11:27 CDT

Arneh Babakhani wrote:
> Hi ,
>
> I have an odd-ball question . . . .
>
> Can NAMD be used to model a general many-body newtonian system? Say
> for instance, a collection of hard spheres moving around and sometimes
> colliding with each other, in a confined space . . . and other than
> collisions, these spheres have no interaction with each other (no
> electrostatics).
> I would think yes . . . you just turn off all of the electrostatic
> (long-range, non-bonded, etc) interactions and parameters, and you
> should be left with a program that just computes F=ma dynamics for
> your system (with temp. and pressure coupling possibly) . . . . no??????
>
> Thanks,
>
> Arneh
>
>
hard spheres don't really mesh well with

f = - grad V

which forms the basis of most of MD. rather, you typically simulate them
via collision detection and (subsequently) momentum reversal. i.e. you
don't have forces in the usual sense. moreover, without nonbonded
forces, the typical expressions for the virial (and hence the pressure)
don't really make all that much sense anymore. none of this should be a
huge deal to implement, but it's quite different from ye olde leapfrog.
So I'd be surprised if it's implemented in NAMD. Take this with a grain
of salt, though, since I'm new to NAMD.

I suppose the answer to your question depends on exactly what you mean
by a "general many-body newtonian system". ;-)

regards,
v.


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:20:28 CST