AMBER Archive (2003)

Subject: Re: AMBER: machine dependent?

From: Shuang Ding (sd517_at_nyu.edu)
Date: Mon Aug 11 2003 - 17:01:38 CDT

Dear Dr. David,

I use Amber 7 in which vlimit default is 20. I tried time step dt=2fs and 1fs, but both has similar problem. And I use 2 processors of OCTANE, and 4 processors of ORIGIN 300. I wonder why different machine get different result.
thanks a lot for your reply.

----- Original Message -----
From: "David A. Case" <case_at_scripps.edu>
Date: Monday, August 11, 2003 4:46 pm
Subject: Re: AMBER: machine dependent?

> On Mon, Aug 11, 2003, Shuang Ding wrote:
> >
> > I'm running a MD simulation of 11mer DNA covalently modified by
> a carcinogen
> > in Amber7. I get a 2.5ns trajectory on our machine OCTANE which
> has 2 250
> > MHZ IP30 Processors and R10000 Chip CPU. But when I use the same
> protocol to
> > run this DNA on our machine ORIGIN 300 with 8 600MHZ IP35
> Processors and
> > R1400 Chip CPU, the simulation stops with SHAKE error at around
> 500ps. And
> > many repeated runs on both machines show much greater stability
> on the
> > OCTANE, with SHAKE errors much less frequent. Is MD simulation
> dependent on
> > different types of machine?
>
> You should (almost) never get shake failures (on any machine) if
> you have
> vlimit set (please check that). See if the repeated shake
> failures have to
> do with atoms in your covalent modification...it may be you are
> getting strong
> forces there (from occasional close contacts). What time step are
> you using?
> Also, it's not clear how many processors you were actually using
> for these
> jobs.
>
> ..regards....dac
>
> --
>
> ==================================================================
> David A. Case | e-mail: case_at_scripps.edu
> Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896
> The Scripps Research Institute | phone: +1-858-784-9768
> 10550 N. Torrey Pines Rd. | home page:
> La Jolla CA 92037 USA | http://www.scripps.edu/case
> ==================================================================
> -------------------------------------------------------------------
> ----
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu