[gmx-users] Continuous mdrun vs step-by-step mdrun

[Venkat Reddy]

Dear gromacs users,

I have a very basic doubt regarding mdrun. Is there any difference between
"doing final MD for 100 ns at  a stretch" and "doing the same with a 10 ns
step size (*i.e., 10ns....20ns....30ns........100ns*) " on a cluster of 256
processors. I have read some where that continuous MD of longer simulations
will cause spurious drifts in velocity and energy, errors in velocity
correlation....etc. Please advise me in this regard.

Thank you and Happy DIWALI

-- 
With Best Wishes
Venkat Reddy Chirasani
PhD student
Laboratory of Computational Biophysics
Department of Biotechnology
IIT Madras
Chennai
INDIA-600036