Re: [AMBER] about amoeba

From: Jason Swails <jason.swails.gmail.com>
Date: Tue, 30 Sep 2014 07:28:23 -0400

On Tue, Sep 30, 2014 at 4:04 AM, m13654940810 <m13654940810.163.com> wrote:

> Hello,
>
> I run simulation with amoeba force field using pmemd.amoeba.MPI,and there
> is a error "Too many diple_dipole interactions for allocated".
>

​It looks like pmemd.cuda underestimated the number of dipole-dipole
interactions that it would need to store. I haven't seen this happen
before. What I can suggest right now is to increase the value of "safety"
in $AMBERHOME/src/pmemd.amoeba/src/amoeba_direct.F90 (I would do from 2.d0
to 3.d0). Then, of course, you need to recompile.

​HTH,
Jason​

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 30 2014 - 04:30:03 PDT
Custom Search