Hi,
> Thomas: can you redo the softcore tests to run 20 steps instead of 100?
> That
> should still provide useful tests, but remove failures arising from drift.
> And, what is going on with the SC_DERIV output??
Fixed. I shortened the test cases a bit. The SC_DERIV problem was
introduced in rev. 10.9 of runmin.f which added a final force call without
the accompanying softcore calculation parts. This way, SC contributions
weren't calculated in the final step and thus missing in the output. All
softcore tests pass fine on my machine now.
However, in the general test suite, I had to comment out the 4096wat_oct
and pb_ivcap0 tests, since the corresponding directories apparently dont
exist...
Kind Regards,
Thomas
Dr. Thomas Steinbrecher
BioMaps Institute
Rutgers University
610 Taylor Rd.
Piscataway, NJ 08854
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Mar 05 2010 - 03:30:03 PST
