On Fri, Jun 10, 2011, hanlu0366 wrote:
> I am sure that my sander is '~/amber11/bin/sander ' . The same errors
> happen when the value of cut changed is 8.
>
> APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
> using 5000.0 points per unit in tabled values
> TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
> | CHECK switch(x): max rel err = 0.2738E-14 at 2.422500
> | CHECK d/dx switch(x): max rel err = 0.8332E-11 at 2.782960
> ---------------------------------------------------
>
> * NB pairs 1385 6666506 exceeds capacity ( 6666666)
> SIZE OF NONBOND LIST = 6666666
> SANDER BOMB in subroutine nonbond_list
> Non bond list overflow!
> check MAXPR in locmem.f
OK: you crystal is probably slightly denser than typical biomolecules, and the
estimate of the amount of space for the nonbonded list must be too small.
In locmem.f, find these lines:
! --- crude (but useful?) estimate for MAXPR:
! DAN ROE: Does this need to be changed for hybridgb
if( igb /= 0 .or. ipb /= 0 ) then
maxpr = 1
else
if( numextra == 0 ) then
maxpr_float = natom * (cutoffnb + skinnb)**3 / 3.0d0 <====
else ! need more nonbon storage with extra points
maxpr_float = natom * (cutoffnb + skinnb)**3 / 2.5d0
end if
and change the "3.0d0" by the arrow above to 2.d0. You should continue to use
a value in the range of 8-10 for "cut".
...good luck...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jun 11 2011 - 07:30:02 PDT
