Hi,
OK, so you have a stable simulation and the box corrections are caused by
"diffusion"
of the box shape. This diffusion issue has been solved for Gromacs 4.0, where
with
isotropic scaling the box shape is fixed.
But the actual problem causing the crashes has not been found yet.
I tried with a simple test system, but could not reproduce the error.
Could you provide me with a tpr file that crashes after a small amount of steps?
This should be possible by running a simulation that crashes with
nstxout,nstvout
and nstenergy relatively high and then using tpbconv.
Berk.
> Date: Thu, 27 Mar 2008 18:50:16 +0100
> From: [EMAIL PROTECTED]
> To: [email protected]
> Subject: [gmx-users] Skewed box using Parrinello-Rahman pressure coupling
>
> Hi Berk,
>
> my volume fluctuations are relative small. For the largest and smallest
> system respectively that I have been simulating the statistics are:
>
> Average RMSD Fluct.
> 207.7 nm^3 0.699 nm^3 0.698 nm^3
> 75.1 nm^3 0.431 nm^3 0.432 nm^3
>
> My pressure coupling parameters are:
>
> ; Pressure coupling
> Pcoupl = Parrinello-Rahman
> pcoupltype = isotropic
> ; Time constant (ps), compressibility (1/bar) and reference P (bar)
> tau_p = 5.0
> compressibility = 5e-5
> ref_p = 1.0
>
>
> Best
>
> Luther
>
> > Hi,
> >
> > I am also starting to wonder now if you might have other problems.
> > Have you checked if there aren't very strong volume fluctuations?
> >
> > Could you mail your mdp pressure coupling parameters?
> >
> > Berk.
> >
> >
> >> Date: Thu, 27 Mar 2008 15:16:39 +0100
> >> From: [EMAIL PROTECTED]
> >> To: [email protected]
> >> Subject: [gmx-users] Skewed box using Parrinello-Rahman pressure
> >> coupling
> >>
> >> Hi Berk,
> >>
> >> I reran my simulations with the settings you suggested. Unfortunately
> >> most
> >> of them crashed. In some of them I received the original error message:
> >>
> >> Source code file: ns.c, line: 313
> >>
> >> Fatal error:
> >> Could not correct too skewed box
> >>
> >> I did not, however, receive this message from all runs. In some cases
> >> conformations were written out for a few steps shortly before the run
> >> crashed, but the simulation process was killed before any error could be
> >> written out. The log-files no proper ending in these cases. Is there any
> >> way to get the error message before the simulation process is killed?
> >>
> >> In one particular simulation the end of the log-file looked like this:
> >>
> >> Correcting invalid box:
> >> old box (3x3):
> >> old box[ 0]={ 5.62033e+00, 0.00000e+00, 0.00000e+00}
> >> old box[ 1]={ 2.79637e+00, 5.26160e+00, 0.00000e+00}
> >> old box[ 2]={-2.81298e+00, -3.13776e-03, 4.54280e+00}
> >> new box (3x3):
> >> new box[ 0]={ 5.62033e+00, 0.00000e+00, 0.00000e+00}
> >> new box[ 1]={ 2.79637e+00, 5.26160e+00, 0.00000e+00}
> >> new box[ 2]={ 2.80735e+00, -3.13776e-03, 4.54280e+00}
> >>
> >> t = 34154.008 ps: Water molecule starting at atom 1615 can not be
> >> settled.
> >> Check for bad cont
> >>
> >>
> >> I am not sure whether there is another problem with this simulation
> >> causing the crash. But since the crash occurred directly after the
> >> correction of the box I thought I should report this as well.
> >>
> >>
> >> Best
> >>
> >> Luther
>
>
> _______________________________________________
> gmx-users mailing list [email protected]
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to [EMAIL PROTECTED]
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/_______________________________________________
gmx-users mailing list [email protected]
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php
