I believe that something IS off when the output is parsed by sander. Our
calculations (on the energy) are WAY off. That's how I came across the
errors. The data from one node was correct and the output of the other was
off.
On Fri, Mar 13, 2015 at 3:27 PM, Jason Swails <jason.swails.gmail.com>
wrote:
> On Fri, 2015-03-13 at 13:54 -0400, Jonathan Gough wrote:
> > Attached is the .diff file.
>
> Oh. I take that back. These diffs look like round-off issues with the
> CHARMM charges (the multiplication factor CHARMM uses is slightly
> different than the one that Amber uses).
>
> These diffs look fine to me. Admittedly, ParmEd uses a method to
> convert the Amber scaling to CHARMM scaling that can lose precision, but
> the diffs you're seeing will have no measurable effect on simulation.
>
> All the best,
> Jason
>
> >
> > On Fri, Mar 13, 2015 at 1:52 PM, Jonathan Gough <
> jonathan.d.gough.gmail.com>
> > wrote:
> >
> > > Hi everyone,
> > >
> > > I am getting 26 errors when I compile AmberTool14:
> > >
> > >
> > > possible FAILURE: check prmtop.new_type.dif
> > > possible FAILURE: check prmtop.NBFIX.dif
> > > possible FAILURE: check prmtop.new_chg.dif
> > > possible FAILURE: check final.prmtop.dif
> > > possible FAILURE: check md3_cpout.pH_4.50.dif
> > > possible FAILURE: check pH_2.00.calcpka.dif
> > > possible FAILURE: check pH_2.50.calcpka.dif
> > > possible FAILURE: check pH_3.00.calcpka.dif
> > > possible FAILURE: check pH_3.50.calcpka.dif
> > > possible FAILURE: check pH_4.00.calcpka.dif
> > > possible FAILURE: check pH_4.50.calcpka.dif
> > > possible FAILURE: check pH_6.00.calcpka.dif
> > > possible FAILURE: check pH_7.50.calcpka.dif
> > > possible FAILURE: check pH_2.00.calcpka.dif
> > > possible FAILURE: check running_avgs.dat.dif
> > > possible FAILURE: check running_avg.dat.dif
> > > possible FAILURE: check running_pKa.dat.dif
> > > possible FAILURE: check chunk.dat.dif
> > > possible FAILURE: check chunk_deprot.dat.dif
> > > possible FAILURE: check chunk_pka.dat.dif
> > > possible FAILURE: check cumulative.dat.dif
> > > possible FAILURE: check cumulative_deprot.dat.dif
> > > possible FAILURE: check cumulative_pka.dat.dif
> > > possible FAILURE: check pop.dat.dif
> > > possible FAILURE: check conditional.dat.dif
> > > possible FAILURE: check conditional_chunks.dat.dif
> > >
> > >
> > > The errors in the .diff file aren't "that big" (I could be wrong) but
> > > unfortunate we are seeing some major differences in the calculations
> we are
> > > doing.
> > >
> > > I didn't write the code that does the calculations, but I am 99.9% sure
> > > that it is the errors in...
> > >
> > > possible FAILURE: check prmtop.new_type.dif
> > > possible FAILURE: check prmtop.NBFIX.dif
> > > possible FAILURE: check prmtop.new_chg.dif
> > > possible FAILURE: check final.prmtop.dif
> > >
> > > that are causing the problems. essentially we are doing a short
> simulation
> > > on a single frame with and without a single water and comparing the
> > > difference in energy.
> > >
> > > we have 2 machines that are essentially identical. one compiles both
> Amber
> > > and AT perfectly and the other does not... I check the gfortran
> versions
> > > and the g++ versions and they are the same.. not sure what else I
> should
> > > check.
> > >
> > > Both versions have the latest updates for Amber and AmberTools:
> > >
> > > AmberTools version 14.26
> > > Amber version 14.10
> > >
> > > If you need more info, please ask. Thanks.
> > > Jonathan
> > >
> > >
> > >
> > >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Mar 13 2015 - 12:30:05 PDT
