Re: [AMBER] vlimit=10 compromise for Amber 20 error: "an illegal memory access was encountered launching kernel kClearForces"?

From: David A Case <david.case.rutgers.edu>
Date: Thu, 15 Oct 2020 14:32:41 -0400

On Thu, Oct 15, 2020, Liao wrote:

>Thanks there, I could try preparing that, though for using vlimit=10,
>does this seem like an appropriate solution? What’s the unit of this
>parameter, the manual doesn’t say it. The trajectory as viewed and post
>MD-MMGbsa energy from it looks normal to me.

Units of vlimit are internal velocity usits: Å/(20 fs). vlimit=10 is
OK, especially if it only get's triggered occasionally. But it does
look like something in your ligand allows large forces: do you have any
hydrogens with zero vdW radii? (This could happen on a ligand if you
were, for example, using gaff rather than gaff2.) Hydrogens are a
common source of headaches for non-standard residues.

....dac


_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Oct 15 2020 - 12:00:02 PDT
Custom Search