Re: [AMBER] freezing bonds in parallel version

From: case <>
Date: Mon, 22 Nov 2010 11:30:20 -0500

On Fri, Nov 19, 2010, Stefano Pieraccini wrote:
> I noticed that it is possible to freeze all the bonds around their
> equilibrium distance when you use the serial version of sander and pmemd,
> while only bonds containing hydrogen can be constrained with shake using
> the parallel version. I know that other software like gromacs and namd allow
> to freeze all bonds even in the parallel version. I wonder which is the
> reason for this different choices,

It simplifies the bookkeeping, but making it easy to ensure that SHAKEn
bonds are always on the same thread. It would be a fair amount of work (but
not outrageous) to remove this restriction.


AMBER mailing list
Received on Mon Nov 22 2010 - 09:00:03 PST
Custom Search