Re: cgmd stability

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Feb 19 2009 - 07:16:46 CST

Generally one needs to initially relax the system with 5 fs timesteps
for the first ns or so, and then can run 10-20 fs timesteps.
Peter

wang wrote:
> Dear all,
> I used the rbcg-2009.par and rbcg-2009.par to simulate my system
> including 300 AAs and 5000 TIP3 .
> When the timestep was great than 10 (fs/step), my system was broken and
> the error message(the atom moving
> too fast ......) was appeared. Then I changed the timestep value and the
> value was equal to 5. And my system is still
> running and stable. Anyone could tell me how the limit of timestep with
> rbcg-2009 parameters ?
> Thanks
> Y.T Wang
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:31 CST