<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#ffffff">
On 19/01/2011 2:48 PM, Cara Kreck wrote:
<blockquote cite="mid:BAY156-w64C4681B38BF751B00594DD4F60@phx.gbl"
type="cite">
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
Hi everyone,<br>
<br>
I have been trying to reproduce a simulation that I previously ran
successfully with gromacs 3.3.3 in gromacs 4.0.7 and 4.5.1. It is
of a tip4p water box generated with Packmol (and no I can't use
the already provided tip4p water box). However I can't even make
it past the energy minimisation step. I have noticed that all
simulations run into trouble at step 12 and 3.3.3 manages to
recover, whereas 4.0.7 and 4.5.1 get stuck. The results of 4.0.7
and 4.5.1 are identical.<br>
<br>
Here's an extract from the 4.0.7/4.5.1 log file:<br>
<br>
<blockquote> Step Time Lambda<br>
11 11.00000 0.00000<br>
<br>
Energies (kJ/mol)<br>
LJ (SR) Coulomb (SR) Coul. recip. Potential
Pressure (bar)<br>
8.12398e+03 -4.64002e+03 -8.80558e+02 2.60340e+03
-4.27109e+03<br>
<br>
Step Time Lambda<br>
12 12.00000 0.00000<br>
<br>
Energies (kJ/mol)<br>
LJ (SR) Coulomb (SR) Coul. recip. Potential
Pressure (bar)<br>
5.67868e+03 -7.67528e+03 -1.19351e+03 -3.19011e+03
-5.97596e+04<br>
<br>
Step Time Lambda<br>
13 13.00000 0.00000<br>
<br>
Step Time Lambda<br>
14 14.00000 0.00000<br>
<br>
Step Time Lambda<br>
15 15.00000 0.00000<br>
<br>
Step Time Lambda<br>
16 16.00000 0.00000<br>
<br>
Step Time Lambda<br>
17 17.00000 0.00000<br>
<br>
Step Time Lambda<br>
18 18.00000 0.00000<br>
<br>
Step Time Lambda<br>
19 19.00000 0.00000<br>
<br>
Step Time Lambda<br>
20 20.00000 0.00000<br>
<br>
Step Time Lambda<br>
21 21.00000 0.00000<br>
<br>
Step Time Lambda<br>
22 22.00000 0.00000<br>
<br>
Step Time Lambda<br>
23 23.00000 0.00000<br>
<br>
Step Time Lambda<br>
24 24.00000 0.00000<br>
<br>
Step Time Lambda<br>
25 25.00000 0.00000<br>
<br>
Step Time Lambda<br>
26 26.00000 0.00000<br>
<br>
Step Time Lambda<br>
27 27.00000 0.00000<br>
<br>
Step Time Lambda<br>
28 28.00000 0.00000<br>
<br>
Step Time Lambda<br>
29 29.00000 0.00000<br>
<br>
<br>
Stepsize too small, or no change in energy.<br>
Converged to machine precision,<br>
but not to the requested precision Fmax < 100<br>
<br>
Double precision normally gives you higher accuracy.<br>
You might need to increase your constraint accuracy, or turn<br>
off constraints alltogether (set constraints = none in mdp file)<br>
<br>
Steepest Descents converged to machine precision in 30 steps,<br>
but did not reach the requested Fmax < 100.<br>
Potential Energy = -3.1901118e+03<br>
Maximum force = 6.9292950e+05 on atom 97<br>
Norm of force = 2.1575133e+04<br>
</blockquote>
<br>
<br>
And from the 3.3.3 log file:<br>
<br>
<blockquote> Step Time Lambda<br>
11 11.00000 0.00000<br>
<br>
Energies (kJ/mol)<br>
LJ (SR) Coulomb (SR) Coul. recip. Potential
Kinetic En.<br>
8.28949e+03 -4.56483e+03 -8.69953e+02 2.85470e+03
0.00000e+00<br>
Total Energy Temperature Pressure (bar)<br>
2.85470e+03 0.00000e+00 0.00000e+00<br>
<br>
<br>
t = 0.012 ps: Water molecule starting at atom 97 can not be
settled.<br>
Check for bad contacts and/or reduce the timestep.Wrote pdb
files with previous and current coordinates<br>
Step Time Lambda<br>
12 12.00000 0.00000<br>
<br>
Step Time Lambda<br>
13 13.00000 0.00000<br>
<br>
Energies (kJ/mol)<br>
LJ (SR) Coulomb (SR) Coul. recip. Potential
Kinetic En.<br>
6.60039e+03 -6.11407e+03 -1.08723e+03 -6.00910e+02
0.00000e+00<br>
Total Energy Temperature Pressure (bar)<br>
-6.00910e+02 0.00000e+00 0.00000e+00<br>
<br>
...<br>
<br>
(continued on with a few more similar warnings but eventually
settled)<br>
<br>
...<br>
<br>
Steepest Descents converged to Fmax < 100 in 233 steps<br>
Potential Energy = -2.3777645e+04<br>
Maximum force = 9.8486107e+01 on atom 1393<br>
Norm of force = 5.2092145e+02<br>
</blockquote>
<br>
<br>
And here's the input parameters:<br>
<br>
<blockquote>cpp = /usr/bin/cpp<br>
define = <br>
constraints = all-angles<br>
unconstrained_start = no<br>
constraint_algorithm = shake<br>
shake_tol = 0.0001<br>
morse = no<br>
integrator = steep<br>
nsteps = 1000<br>
; Energy minimizing <br>
emtol = 100<br>
emstep = 0.01 <br>
nstcomm = 1<br>
nstlist = 10<br>
ns_type = grid<br>
pbc = xyz<br>
rlist = 0.9<br>
coulombtype = pme<br>
rcoulomb = 0.9<br>
vdw-type = cut-off<br>
rvdw = 0.9<br>
nstenergy = 10<br>
Tcoupl = no<br>
Pcoupl = no<br>
gen_vel = no<br>
</blockquote>
<br>
<br>
I then tried running the same parameters using the spc water box
in the tutorial and got these results:<br>
<br>
4.0.5/4.7.1<br>
<br>
<blockquote>Steepest Descents converged to Fmax < 100 in 63
steps<br>
Potential Energy = -1.0896633e+04<br>
Maximum force = 9.1075676e+01 on atom 202<br>
Norm of force = 1.3547868e+01<br>
</blockquote>
<br>
3.3.3<br>
<br>
<blockquote>
Steepest Descents converged to Fmax < 100 in 32 steps<br>
Potential Energy = -1.0760295e+04<br>
Maximum force = 8.2715759e+01 on atom 425<br>
Norm of force = 5.6500586e+02<br>
<br>
</blockquote>
<br>
It also gave the same result for version 4 if I ran 4 with the
.tpr file I had prepared using 3.<br>
<br>
I am running these remotely and I didn't know whether single or
double precision versions were installed. The .trr file from 3.3.3
was twice the size of the others (even with half the steps), but
gmxdump told me that all versions were single precision.<br>
<br>
Are the differences simply due changes between the versions?
Looking at the release notes, most changes (particularly between
3.3.3 and 4.0.7) were related to parallelisation and I've run
everything on a single cpu. Or is there something else I'm
missing? <br>
</blockquote>
<br>
Of itself, I don't think this is a problem. The same software with a
different compiler can change its numerical stability, in extreme
cases. It would be surprising for 4.0 to have introduced an
algorithmic change that breaks EM, and for it to be first noticed
now. It is normal not to run EM with constraints, and I suspect some
change in the constraint implementation may be the origin of your
observations. Try without constraints.<br>
<br>
Mark<br>
</body>
</html>