AMBER Archive (2001)

Subject: Re: SGI_MP (was "your mail")

From: Nathalie Godbout (godbout_at_sgihud.hudson.sgi.com)
Date: Fri Mar 23 2001 - 13:54:14 CST

Jarrod Smith wrote:
>
> Nathalie Godbout wrote:
> > Hi,
> >
> > There is a source code error in sander_classic that I've reported and
> > which is documented on the Amber web site:
> >
> > http://www.amber.ucsf.edu/amber/bugfixes/6.0/bugfix.20
> >
>
> Nathalie,
>
> If we are running gibbs using shared memory parallel, should we
> compile it in a similar way as well, or is this only an issue with the
> qiktip.o object in sander_classic?
>
> Thanks,
>
> Jarrod A. Smith
> Research Asst. Professor, Biochemistry
> Asst. Director, Center for Structural Biology
> Computation and Molecular Graphics
> Vanderbilt University
>
> jsmith_at_structbio.vanderbilt.edu

Hi Jarrod,

Sorry for the delay, I had to check a few things in the source code. The
answer is it's only an issue with sander_classic. The subroutine qiktip
in gibbs is different to that in sander_classic. The bug is in the part
which is different. You would have noticed a problem because this bug
leads to `nan' energy values for parallel runs.

Nathalie 

-- 
------------------------------------------------------------------------
Nathalie Godbout, Ph.D.       SGI 
godbout_at_hudson.sgi.com        Chemistry and Biology Applications Group
(781)839-2147                 One Cabot Road, Suite 250
(978)562-7450 fax             Hudson, MA     
vnet:483-2147                 01749, USA
------------------------------------------------------------------------