On Thu, Dec 18, 2014, Josh Berryman wrote:
>
> Do we need to
> keep supporting the old (multiple parmtop) way of doing it?
The sander approach supports things like TI for QM/MM (e.g. V0 is QM/MM,
V1 is MM only) that would rather difficult to implement with the single
prmtop approach. It also supports experimentation for other things where
one is moving between two very different types of Hamiltonians.
But I'm in favor of simplification, and open to arguments about what is best.
My mdread2() change is not a "fix", just a documentation of a current
limitation in the code. It should be pretty easy to separate the logdvdl
functionality from the softcore functionality.
....dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Dec 18 2014 - 05:30:05 PST
