<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
FONT-SIZE: 10pt;
FONT-FAMILY:Tahoma
}
</style>
</head>
<body class='hmmessage'><div style="text-align: left;">Hi,<br><br>I have not heard of any code problems with pressure coupling.<br></div>But in general things can get unstable when you couple to quickly.<br>0.5 ps is quite short, but should be ok for isotropic coupling in water.<br>For semiistropic coupling it might actually be too fast, I have never<br>tried this. Semiisotropic coupling will lead to larger and faster<br>fluctuations.<br><br>For the large system I would set tau_p to somewhere between 1 and 5 fs.<br><br>I don't know if you compressibility value is reasonable when you have<br>a membrane in it, but I guess it would not be to far off.<br><br>One last thing, don't you want to use surface tension coupling?<br><br>Berk.<br><br><hr id="stopSpelling">> Date: Wed, 4 Jun 2008 11:54:53 +0200<br>> From: till@bph.rub.de<br>> To: gmx-users@gromacs.org<br>> Subject: Re: [gmx-users] neighborsearching in semiisotropic systems<br>> <br>> Hi Berk,<br>> the whole system I want to simulate contents a membrane (that's why I <br>> use semiisotropic pressure coupling) and this system crashed with the <br>> same error message.<br>> In the membrane system and in the small test system no distortion of the <br>> box can be observed during the simulation. The cubic form of the box is <br>> stable.<br>> <br>> <br>> Step Time Lambda<br>> 1555000 3110.00000 0.00000<br>> <br>> Rel. Constraint Deviation: Max between atoms RMS<br>> Before LINCS 0.016766 46 47 0.003642<br>> After LINCS 0.000042 23 25 0.000013<br>> <br>> Energies (kJ/mol)<br>> G96Angle Proper Dih. Improper Dih. LJ-14 Coulomb-14<br>> 1.61040e+02 6.05084e+01 3.35858e+01 1.03030e+01 1.32855e+03<br>> LJ (SR) LJ (LR) Coulomb (SR) Coul. recip. Potential<br>> 2.25304e+04 -6.11549e+02 -1.40294e+05 -1.72589e+04 -1.34040e+05<br>> Kinetic En. Total Energy Temperature Pressure (bar)<br>> 2.39730e+04 -1.10067e+05 2.97751e+02 -1.18958e+02<br>> <br>> Grid: 6 x 6 x 8 cells<br>> Grid: 5 x 5 x 8 cells<br>> Grid: 6 x 6 x 8 cells<br>> Grid: 5 x 5 x 8 cells<br>> Grid: 6 x 6 x 8 cells<br>> Grid: 5 x 5 x 8 cells<br>> Grid: 6 x 6 x 8 cells<br>> Grid: 5 x 5 x 8 cells<br>> -------------------------------------------------------<br>> Program mdrun_d.mpi, VERSION 3.3.1<br>> Source code file: nsgrid.c, line: 226<br>> <br>> Range checking error:<br>> Explanation: During neighborsearching, we assign each particle to a grid<br>> based on its coordinates. If your system contains collisions or parameter<br>> errors that give particles very high velocities you might end up with some<br>> coordinates being +-Infinity or NaN (not-a-number). Obviously, we cannot<br>> put these on a grid, so this is usually where we detect those errors.<br>> Make sure your system is properly energy-minimized and that the potential<br>> energy seems reasonable before trying again.<br>> <br>> Variable ci has value 284. It should have been within [ 0 .. 200 ]<br>> Please report this to the mailing list (gmx-users@gromacs.org)<br>> -------------------------------------------------------<br>> <br>> <br>> I have no explanation why the system with semiisotropic pressure <br>> coupling crashes within the 10ns simulation with tau_p=0.5 and does not <br>> crash with tau_p=5.0. The other problem is starting two identical <br>> simulations the systems crashed at different timesteps.<br>> <br>> Till<br>> <br>> Berk Hess schrieb:<br>> > Hi,<br>> > <br>> > Did you check how your box dimensions evolve?<br>> > <br>> > For a homogeneous system (as your small test system)<br>> > one should not use semi or anisotropic pressure coupling,<br>> > since the box shape can distort unlimitedly causing<br>> > the system to crash.<br>> > <br>> > Berk.<br>> > <br>> > <br>> > ------------------------------------------------------------------------<br>> > > Date: Tue, 3 Jun 2008 17:00:14 +0200<br>> > > From: till@bph.rub.de<br>> > > To: gmx-users@gromacs.org<br>> > > Subject: [gmx-users] neighborsearching in semiisotropic systems<br>> > ><br>> > > Dear users,<br>> > ><br>> > > During protein simulations occurred the range checking error of the<br>> > > neighborsearching. I checked the comments on this mistake in the user<br>> > > list and then checked my system on any hints of an exploding system but<br>> > > it seems to be stable. In order to be sure I cut down the system to a<br>> > > small peptide consisting of only ten aminoacids and used the ffG43a1<br>> > > forcefield.<br>> > ><br>> > ><br>> > > I performed 10ns simulations of this small system in a cubic<br>> > > spc-waterbox with different pressure parameters and tried single and<br>> > > double precision. I got the following results:<br>> > ><br>> > ><br>> > > For<br>> > ><br>> > > pcoupltype = semiisotropic<br>> > ><br>> > > tau_p = 0.5 0.5<br>> > ><br>> > > compressibility = 4.5E-5 4.5E-5<br>> > ><br>> > > ref_p = 1 1<br>> > ><br>> > ><br>> > > with double precision occurred the mistake:<br>> > ><br>> > ><br>> > > -------------------------------------------------------<br>> > ><br>> > > Program mdrun_d.mpi, VERSION 3.3.1<br>> > ><br>> > > Source code file: nsgrid.c, line: 226<br>> > ><br>> > ><br>> > > Range checking error:<br>> > ><br>> > > Explanation: During neighborsearching, we assign each particle to a grid<br>> > ><br>> > > based on its coordinates. If your system contains collisions or parameter<br>> > ><br>> > > errors that give particles very high velocities you might end up with <br>> > some<br>> > ><br>> > > coordinates being +-Infinity or NaN (not-a-number). Obviously, we cannot<br>> > ><br>> > > put these on a grid, so this is usually where we detect those errors.<br>> > ><br>> > > Make sure your system is properly energy-minimized and that the potential<br>> > ><br>> > > energy seems reasonable before trying again.<br>> > ><br>> > ><br>> > > Variable ci has value 284. It should have been within [ 0 .. 200 ]<br>> > ><br>> > > Please report this to the mailing list (gmx-users@gromacs.org<br>> > > <mailto:gmx-users@gromacs.org>)<br>> > ><br>> > ><br>> > > Then, I changed tau_p to 5.0 and the mistake did not occurce anymore<br>> > > during the 10 ns. With single precision the mistake never occurres. I<br>> > > also tested double precision with pcoupltype = isotropic and this also<br>> > > works.<br>> > ><br>> > > Next, I wanted to reproduce the mistake so I took the same tpr file of<br>> > > the crashed simulation with tau_p = 0.5 and started the simulation<br>> > > again. The mistake occurred again but not at the same timestep. The<br>> > > first run crashed after step 1558630 and the second run after step<br>> > > 2597500 and also the ci variable is not exactly the same (Variable ci<br>> > > has value 283. It should have been within [ 0 .. 200 ]).<br>> > ><br>> > ><br>> > > I think there might be a problem between the neighborsearching and the<br>> > > semiisotropic pressurcoupling together with double precision. I am quite<br>> > > sure that the peptide is not the problem. Maybe there is a problem with<br>> > > a water molecule. I would be glad if someone can give me an explanation<br>> > > why this mistake is not reproducable at the same timestep and why it<br>> > > occures with tau_p=0.5 and not with tau_p=5.0.<br>> > ><br>> > ><br>> > > Thanks<br>> > ><br>> > ><br>> > > Till<br>> > ><br>> > > _______________________________________________<br>> > > gmx-users mailing list gmx-users@gromacs.org<br>> > > http://www.gromacs.org/mailman/listinfo/gmx-users<br>> > > Please search the archive at http://www.gromacs.org/search before <br>> > posting!<br>> > > Please don't post (un)subscribe requests to the list. Use the<br>> > > www interface or send it to gmx-users-request@gromacs.org.<br>> > > Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br>> > <br>> > ------------------------------------------------------------------------<br>> > Express yourself instantly with MSN Messenger! MSN Messenger <br>> > <http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/><br>> > <br>> > <br>> > ------------------------------------------------------------------------<br>> > <br>> > _______________________________________________<br>> > gmx-users mailing list gmx-users@gromacs.org<br>> > http://www.gromacs.org/mailman/listinfo/gmx-users<br>> > Please search the archive at http://www.gromacs.org/search before posting!<br>> > Please don't post (un)subscribe requests to the list. Use the <br>> > www interface or send it to gmx-users-request@gromacs.org.<br>> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br>> _______________________________________________<br>> gmx-users mailing list gmx-users@gromacs.org<br>> http://www.gromacs.org/mailman/listinfo/gmx-users<br>> Please search the archive at http://www.gromacs.org/search before posting!<br>> Please don't post (un)subscribe requests to the list. Use the <br>> www interface or send it to gmx-users-request@gromacs.org.<br>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br><br /><hr />Express yourself instantly with MSN Messenger! <a href='http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/' target='_new'>MSN Messenger</a></body>
</html>