AMBER Archive (2009)
Subject: Re: [AMBER] xmin
From: Scott Brozell (sbrozell_at_rci.rutgers.edu)
Date: Mon May 18 2009 - 15:30:17 CDT
(I actually used the development version, amber11.)
On Mon, May 18, 2009 at 01:58:00AM +0100, Peter Varnai wrote:
> I am surprised that you could run test/dhfr/Run.dhfr.min just by
> adding ntmin=3. I had to tinker a bit more to make it run:
> ntx=7, irest=1, ntmin=3,
> ntc=1, ntf=1,
> imin=1, ntb=0, cut=10.,
> ntpr=1, ntwr=10000, maxcyc=500, drms=0.01,
> but it stops early:
> NSTEP ENERGY RMS GMAX NAME NUMBER
> 22 -6.0115E+04 4.2326E+00 3.7434E+01 H1 20157
> BOND = 5167.9651 ANGLE = 717.5174 DIHED = 858.5201
> VDWAALS = 8959.3791 EEL = -82962.3361 HBOND = 0.0000
> 1-4 VDW = 500.2248 1-4 EEL = 6644.0066 RESTRAINT = 0.0000
> On Sun, 17 May 2009, Scott Brozell wrote:
> > On Thu, May 14, 2009 at 10:29:09PM +0100, Peter Varnai wrote:
> > >
> > > I noticed that that xmin minimiser (amber 10) terminates the cycles
> > > before it reaches either maxcyc or drms. It prints RMS=0.0 values
> > > along the cycles except in the 'final results'. It does not seem to be
> > > able to handle periodic systems. Is there a way to force xmin to reach
> > > maxcyc or drms?
> > There is a problem at the least with the RMS.
> > However, using ntmin=3 in test/dhfr/Run.dhfr.min does terminate at maxcyc=5
> > For bug reporting purposes what range of maxcyc or drms have you tried ?
> > It may take a while for me to get to this.
AMBER mailing list