On Wed, Jan 13, 2010, Cihan Aydin wrote:
>
> My AMBER version is a fully-patched 10 and all serial and parallel test
> cases pass. I use sander.MPI to run my processes. I am seeing the
> restart file written at the end of the simulation (in this case,
> minimization) but not during the simulation and no file is created
> before the end.
Assuming(?) that the file you posted is the one that shows the problem:
ntwr only applies to dynamics, not to minimization runs. I guess we never
figured that minimzation runs would be long enough to warrant such an option.
(Even the longish job you show only took two hours.)
In our defense, the manual *is* correct here: "every NTWR steps during
dynamics, the "restart" file will be written...."
>
> Before the files, I also produced another question... I was trying to
> run a xmin (TNCG) on my system (~20000 atoms, 647 residue protein others
> TIP3) but it crashes (core-dumped segmentation fault). Is this system
> too big for non-linear optimization?
No.
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jan 13 2010 - 19:00:02 PST
