Hi, Ross. Thanks for replying with suggestions.
I tested the installations (make test) after serial configuring as well as cuda. For good measure, I removed my .profile and .bashrc so no PYTHONPATH or any other stay environmental variable should mess the process.
For a reinstall (make distclean and then ./configure gnu) I got 204 comparisons passed and no diffs to save.
I rm -rf that installation, untar AmberTools18 and Ambers18 and compiled it making sure every update was in place.
I then ran the tests and got the same result: 204 comparisons passed no diffs to save.
A MD run test (same protein I got the original error) gave the same results:a weird raise in energy during thermal equilibration.
So, any other suggestions?
Best regards.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Nov 21 2018 - 14:30:04 PST
