[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ESPResSo-devel] Nsquare cellsystem and folded positions

From: Olaf Lenz
Subject: Re: [ESPResSo-devel] Nsquare cellsystem and folded positions
Date: Mon, 13 Aug 2007 15:57:03 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv: Gecko/20070728 Thunderbird/ Mnenhy/

Hash: SHA1


Limbach, Hans Joerg, LAUSANNE, NRC-FS wrote:
> Why it doesn't make sense...
> ... Put one particle in a box. Apply periodicity in 1D. 
> Then you have always two images of the original particle
> which are at the very same distance. Which one you choose
> to calculate the force?

For simulations employing short-ranged interactions, the size of the
simulation box has to be at least two times the maximal interaction
range. Then you can safely assume and use the minimum image convention.
The self-images can be safely ignored, as their distance are always
larger than the maximal interaction range.

When you use long-ranged forces, you cannot apply the minimum image
convention. Under these circumstances, neither cell-lists nor
Verlet-lists will help you. You either have to use the direct or Ewald
summation, or P3M, or the like.

> And that's right you can not simulate a system with periodic bounderies 
> (and minimum image convention) where the box is smaller then 2
> times the maximal interaction range.

However, I was talking about box sizes between 2 and 3 times the maximal
interaction range. And that should be possible, alas, it is not with

> Documentation of all this would mean to write a textbook about 
> simulations in general not about Espresso.

No, sorry, I'm not with you in this case. ESPResSo is pretty special
here, as it completely mixes up the concepts of the boundary conditions,
the parallelization scheme, and the algorithmical acceleration (Verlet
lists, cell lists, etc.), which are basically independent.

It is by no means obvious that the nsquare system does not allow for
periodic boundary conditions. In fact, this is probably the most basic
simulation type for a liquid that you can think of!

I think it is definitely necessary to document this limitation of ES, or
 otherwise to remove it.

Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla -


reply via email to

[Prev in Thread] Current Thread [Next in Thread]