You can set ig=-1, and sander/pmemd will select a random seed based on the
current time. It must be pmemd from Amber10 with all bug fixes applied,
though. Otherwise you should be setting ig explicitly each time to avoid
synchronization artifacts like Hirdesh said.
Hope this helps,
Jason
On Fri, Sep 3, 2010 at 4:16 AM, hirdesh kumar <hirdeshs8.gmail.com> wrote:
> Yes. You should set a new ig value on each restart to prevent
> synchronization.
>
> On Fri, Sep 3, 2010 at 1:19 PM, colvin <colvin4367.gmail.com> wrote:
>
> > Dear all,
> >
> > When i was reading the manual for temperature regulation, i saw a
> > statement: "Since Langevin simulations are potentially susceptible to
> > "synchronization" artifacts, you should explicitly set the ig variable
> > (described below) to a different value at each restart of a given
> > simulation."
> >
> > I check amber mailing list for this ig value as well...and in one of
> > the tutorial (Loop Dynamics of the HIV-1 Integrase Core Domain) ig
> > value was used.
> >
> > If i want to use ntt=3 and run the 1.8 ns of simulation by 9x200ps
> > run, as shown by amber workshop tutorial-A Practical Example (A-DNA),
> > do i need to set ig value? Because the tutorial does not set this ig
> > value.
> >
> > in short, all simulations that use ntt=3, regardless the system size
> > and nature, will need to set ig value? same for sander and pmemd?
> >
> > Thanks and regards,
> >
> > Colvin
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 03 2010 - 05:30:03 PDT
