AMBER Archive (2009)

Subject: Re: [AMBER] The -c option in Sander

From: case (
Date: Fri Oct 09 2009 - 20:54:05 CDT

On Fri, Oct 09, 2009, phani ghanakota wrote:
> 1) I would like to know how the -c option in sander works. If we give
> a file after the -c option and specify irest = 0 and NTX = 5, will the
> velocity information be read in?

If you want to continue an MD simulation from a previous restart file, you
should set irest=1.

I don't know what happens if irest=0 and ntx=5; it certainly depends on
other input (especially, but not limited to, tempi). I believe that the
old velocities being read in will not be assigned to the correct time
point if you use the options you suggest.

If you want to find out for sure, either study the code, or run some
empirical tests (e.g. with ntpr=1 and nstlim=10).


AMBER mailing list