Re: reproduciblity between software

From: Mark Abraham (Mark.Abraham_at_anu.edu.au)
Date: Fri Mar 02 2007 - 17:59:17 CST

Stern, Julie wrote:
> Hello,
> Have there been any studies or comparisons done regarding reproducibility of
> an MD result in amber vs. namd? If all the parameter options are set the same and
> the initial conditions are the same, are the algorithms in amber and namd implemented
> the same so that an exact trajectory would come out the same? Where do you suppose
> the differences would be?

There is no chance they would reproduce each other. Computer arithmetic
is only an approximation to the true arithmetic, as described here
http://en.wikipedia.org/wiki/Floating_point#Accuracy_problems. You would
not even expect neighbouring versions of the same software to reproduce
each other. Tiny differences magnify according to the butterfly effect -
MD simulations are chaotic systems.

They should sample the same ensemble however, and in the limit of long
simulations, their observables will be equal.

Mark

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:04 CST