[gmx-users] weird nst*out problem

[Bert de Groot]

Hi all,

we have noticed the weird problem that when the simulation time gets large
compared to nstxout, nstxtcout, or any of the other nst*out parameters,
then the output frequency in the .xtc, or .trr file becomes inaccurate.

Example:

nstxtcout                = 50
tinit                    = 300000

will yield a .xtc, which, when fed to gmxcheck, will give:

Checking file traj.xtc
Reading frame       0 time 300000.000   
# Atoms  648
Precision 0.001 (nm)
Reading frame       3 time 300000.312   
Timesteps at t=300000 don't match (0.09375, 0.125)
Reading frame       4 time 300000.406   
Timesteps at t=300000 don't match (0.125, 0.09375)
Reading frame       8 time 300000.812   
Timesteps at t=300001 don't match (0.09375, 0.125)
Reading frame       9 time 300000.906   
Timesteps at t=300001 don't match (0.125, 0.09375)

etc.

To pin down the problem, we found out that the problem arises when
time / nst*out  > ~5243

So, for example, with nst*out = 50, the problem will arise from 
approximately t=262000 (262 ns), or from t= 524 ns, with nst*out = 100.

Any ideas?

Bert

____________________________________________________________________________
Dr. Bert de Groot

Max Planck Institute for Biophysical Chemistry
Theoretical molecular biophysics group
Am Fassberg 11 
37077 Goettingen, Germany

tel: +49-551-2011306, fax: +49-551-2011089

email: bgroot at gwdg.de
http://www.mpibpc.gwdg.de/abteilungen/071/bgroot
____________________________________________________________________________