[gmx-users] (no subject)

[Mark Abraham]

On 27/09/2011 11:04 PM, Prema Awati wrote:
> Hi,
>       Thanks for your suggestion; I am attaching the vac plot for 
> short time interval ; Please let me know whether its correct or not.
> regards.
>
>

Atoms are colliding over a time scale not that much larger than the 
integration time step (else you'd probably use a larger time step), and 
the velocities change during collisions, so the characteristic times of 
velocity autocorrelation are going to be not that much larger than the 
integration time step. To measure how velocities correlate with 
themselves, you need to make observations smaller than that 
characteristic time. Your graph suggests you collected data every 5ps 
(maybe every 2500 integration steps?), and shows that there is little 
self-correlation at or beyond 5ps, which is what you might have expected.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110928/9235ca76/attachment.html>