Dear Adrian,
This run was launched from regular restart after short temperature scaling
and density equilibration. Step of integration is 2 fs long, and yes, I set
to write every 500th step. Calculations are running now, but suddenly I've
noted this shift in time. Final NETCDF file will be approximately 100 Gb. I
have to store such a large data for a while for my analysis.
2012/1/3 Adrian Roitberg <roitberg.ufl.edu>
> Dear Dmitry
>
> Is this about the 'time' being printed 'wrong' ? Sees to be so. However,
> how could this possibly matter ? The time printed there can be easily
> computed from NSTEP*dt, and in fact is referenced to the first step
> time. You could easily change that by changing the initial restart file
> times stamp at the top.
>
> Second, and perhaps more important, you seem to be writing your mdout
> file every 500 steps ! Is this really needed ? Unless yous system is
> really small, you are probably wasting time doing that and you mdout
> would be HUGE. I would change ntpr to 10000 or even more.
>
> If you are also writing the mdcrd file this often, I am afraid for your
> disk ! ;-)
>
> Adrian
>
>
--
Sincerely,
Dmitry Mukha
Institute of Bioorganic Chemistry, NAS, Minsk, Belarus
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jan 03 2012 - 08:30:02 PST
