Hi,
On Mon, Aug 26, 2013 at 8:48 AM, Jan-Philip Gehrcke
<jgehrcke.googlemail.com> wrote:
> - the min1+min2+heat protocol runs fine with the CPU version, i.e. CPU
> minimization solves the problem. The crash during heatup, however, also
> appears in the CPU version when starting from the minimized structure as
> created by the GPU code.
>
> - it is very interesting to understand what is wrong with the structure
> in min2.rst as created by the GPU version. MOE does not identify
> clashes, the structure looks fine. Any pointers what I can check to
> identify the problematic part in this structure?
The crash corresponds to this message:
> vlimit exceeded for step 3; vmax = 64.6955
>
> Coordinate resetting cannot be accomplished,
> deviation is too large
> iter_cnt, my_bond_idx, i and j are : 2 948 435 434
This is a SHAKE issue (so checking for clashes in the structure
doesn't necessarily help). The bond between atoms 435 and 434 is
probably too large, which may indicate a prior minimization with SHAKE
off was too short. This is probably related to the 'vlimit' issue as
well (huge forces are being generated because the structure is not yet
well-minimized which is in turn accelerating atoms too much). Note
that in the GPU code the 'vlimit' check is not performed I think for
speed reasons, so you would never get that message with the GPU code.
You can try checking the structure with the 'check' command in
cpptraj, which looks for both overlaps and strange bond lengths
(although you may have to play with the default cutoff).
-Dan
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-9119 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 26 2013 - 08:30:02 PDT
