[gmx-users] Longer scale simulations

[Berk Hess]

> > Regarding SD - is it possible to divide the system to a region were full 
>MD takes place and a region where SD is effective?

>we can not combine these.

They can be combined.
You can use different friction constants for different groups of atoms.
A very small friction produces results which are almost indistinguishable 
from MD.
The only problem is that you should not have to much cut-off artifacts, as 
these might
alter the temperature.
But such an approach will not make the simulation faster, as the timestep 
will be limited
by the MD part and unless the friction constant is very high, SD will not 
allow for larger
timesteps then MD.

Berk.

_________________________________________________________________
Talk with your online friends with MSN Messenger http://messenger.msn.nl/