On Fri, Dec 2, 2011 at 10:03 AM, Jorgen Simonsen <jorgen589.gmail.com>wrote:
> Hi
>
> I get the following error from my md simulation
>
> At line 2763 of file _ew_setup.f
> Fortran runtime error: Bad value during integer read
>
This doesn't actually give us enough information. Line numbers in
_ew_setup.f are dependent on the pre-processor flags used during
compilation (i.e. -DMPI for parallel, -DLES for LES, etc.). Looking at the
read commands in ew_setup.f makes me guess that the error comes in trying
to read the number of atoms from the inpcrd file, so check to make sure
that the file you gave to "-c" was *actually* a restart file or inpcrd, and
not a trajectory file or something else (i.e. prmtop) by accident.
The only other place integers are read are in the &ewald namelist, which
you didn't include in your input file (and I think the error message would
be different, anyway).
> I want to start my simulation from a different point in the trajectory
> and I have taken the restart coordinate like this
>
> trajin ../../MD_20_40ns/mdcrd 2168 2168
> center :1-62 mass origin
> image origin center familiar
> trajout lowest_energy_struct.rst rst
>
> and I want to do the simulation with periodic boundary conditions with
> the pressure set to 1 - my input file looks like this
>
> Rst2
> &cntrl
> imin = 0,irest=0,ntx =5,
irest=0 and ntx=5 cannot be used together. You should have seen a message
like "NTX / IREST inconsistency" in your mdout file somewhere. What's
strange is that this isn't actually a fatal error in sander (but it is in
pmemd). Try setting irest=1 to restart a simulation, or ntx=1 to start
from a fresh minimization.
HTH,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Dec 02 2011 - 08:00:03 PST
