Hi Anna,
> Amber test runs - all successful
> Old TI files which already ran successfully in Amber 10 - successful
> TI with the same file for V0 and V1 as suggested by Tom results in the
> same errors as TI with different files.
These three findings, combined with the fact that it still doesn't run are
fairly puzzling.
The only difference in your prmtops is the charge list, right? and both
prmtops can be run on their own as a TI calc with no change.
There is one more thing you can try:
Make a copy of the V0 prmtop and change one charge to the corresponding
value it would have in V1.
Run a TI calc with V0(original) - V0 (copy with one changed charge) to see
if you get the same MPI crash. Dont worry about the non-integer total
charge.
Should that work, change another charge to see if you can make the error
occur by gradually changing V0 into V1.
If it doesnt work, then my ideas are exhausted, except maybe recompile
sander together with a different MPI. Or, if you want, send me your files,
I'll give it a try here.
Thomas
Dr. Thomas Steinbrecher
formerly at the
BioMaps Institute
Rutgers University
610 Taylor Rd.
Piscataway, NJ 08854
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 02 2012 - 03:00:02 PDT
