From: Blake Charlebois (bdc_at_mie.utoronto.ca)
Date: Thu Jun 09 2005 - 11:53:35 CDT
Hi Pijush.
I think the answer is no, there is nothing wrong with restarting with binary
velocity and coordinate restart files. That said, I am not an authority on
MD simulations.
The thread below may be of interest. (N.B., I believe Jim Phillips is the
lead NAMD developer.) There are quite a few related threads.
http://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l/1317.html
Blake
-----Original Message-----
From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf
Of Pijush Ghosh
Sent: June 8, 2005 2:40 PM
To: namd-l_at_ks.uiuc.edu
Subject: namd-l: Number of Steps
Hello
I have a question on the numsteps(time) of simulation.
Instead of running my minimization or MD simulation for time T, if I break
it into t1, t2, t3.... (T>>t), with output from one carried to the next, am
I doing a big approximation or building up a huge error. I want to follow
this to make things computationally convenient.
I am looking forward for some discussion on this.
Any reference of journal paper or book on this is highly appreciated.
Pijush.
Pijush Ghosh
PhD Student
Department of Civil Engineering
North Dakota State University
Fargo. ND. 58105. USA
Phone:
701-231-6491(Lab)
701-231-4341(Res)
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:39:32 CST