This is something that has caused problems for LMOD all along and,
therefore, LMOD uses a simple script to alter the parameter file to
fix this "Ice Nine" behavior.
Istvan
Quoting Ross Walker <ross.rosswalker.co.uk>:
> Hi All,
>
> We are seeing more and more examples on the GPU code of calculations
> crashing due to zero VDW radii on Hydrogens. This is becoming more and
> more of a problem as we move to longer and longer time scales. Please take
> a look at the following bug report:
>
> http://bugzilla.ambermd.org/show_bug.cgi?id=180
>
> Specifically the last comment.
>
> -----
> Just hacked the VDW parameters for the hydrogens to 0.0000001.
>
> Marking this as resolved. We need to fix the force field - the code is
> fine.
> Lachele, if you can find a case where the VDW radii are non-zero, re-open
> the
> bug
> and send it to me. Otherwise, we're done here.
>
> Ross, you could put a warning about this in the FORTRAN code since it's no
> gpu-specific, but I really think the solution is to not entirely zero out
> the
> VDW radii.
>
> -----
>
> So I am proposing that we have code in rdparm that detects zero VDW radii,
> prints a warning message about it and then adjusts these radii in the code
> to 0.0000001.
>
> Objections, comments / other solutions?
>
> Should we do this inside the force field parm files instead? - Or have
> leap write 0.0000001 instead of 0?
>
> All the best
> Ross
>
> /\
> \/
> |\oss Walker
>
> ---------------------------------------------------------
> | Associate Research Professor |
> | San Diego Supercomputer Center |
> | Adjunct Associate Professor |
> | Dept. of Chemistry and Biochemistry |
> | University of California San Diego |
> | NVIDIA Fellow |
> | http://www.rosswalker.co.uk | http://www.wmd-lab.org |
> | Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
> ---------------------------------------------------------
>
> Note: Electronic Mail is not secure, has no guarantee of delivery, may not
> be read every day, and should not be used for urgent or sensitive issues.
>
>
>
>
>
>
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Aug 16 2013 - 11:00:02 PDT
