[gmx-users] Excessive compute time using PME

[Jordi Camps]

Hello,

I started with a EM and a slightly bigger box (5% bigger). But some problem
remains:
	Steepest Descents:
	   Tolerance (Fmax)   =  1.00000e+03
	   Number of steps    =          500
	Fatal error: ci = -2147483648 should be in 0 .. 2001 [FILE nsgrid.c,
LINE 218]

Where is supposed to be the center of the box? Because in the the file, only
the vectors are given, but no reference point.

Thanks!

--

Jordi Camps Puchades

Instituto Nacional de Bioinformatica (INB) Nodo Computacional GNHC-2
UPC-CIRI
c/. Jordi Girona 1-3 
Modul C6-E201                     Tel. : 934 011 650
E-08034 Barcelona                 Fax  : 934 017 014
Catalunya (Spain)                 e-mail: jcamps at lsi.upc.edu 

> -----Mensaje original-----
> De: gmx-users-bounces at gromacs.org 
> [mailto:gmx-users-bounces at gromacs.org] En nombre de David van 
> der Spoel
> Enviado el: jueves, 19 de mayo de 2005 14:31
> Para: Discussion list for GROMACS users
> Asunto: RE: [gmx-users] Excessive compute time using PME
> 
> On Thu, 2005-05-19 at 13:13 +0200, Jordi Camps wrote:
> > Hello,
> > 
> > I have some news about the problem. Yesterday I saw that the input 
> > that I was feeding was not completely contained in the 
> octahedron, so 
> > I recentered the travelling waters and now I have a true 
> octahedron as input for mdrun.
> > If I chech the coordinates of the tpr file (extracted with 
> editconf) I 
> > continue seeing an octahedron in VMD.
> > But when I run the simulation with mdrun, it fails. And before core 
> > dumping, mdrun writes two files: step-1.pdb and step0.pdb.
> > Step0.pdb shows an exploded molecule (and waters).
> > Step-1.pdb shows my input in cubic boundaries. I think that this is 
> > the problem. I'm feeding an octahedron and mdrun is 
> interpreting this 
> > like a box. There are some special considerations I should 
> take into 
> > account if I want to simulate with an octahedron? Perhaps 
> I'm missing 
> > some parameter in the grompp.mdp file? The box vectors, 
> like the ones 
> > that I have in the file, are 6.72730 6.34256 5.49282 
> 0.00000 0.00000 
> > 2.24243 0.00000 -2.24243 3.17128.
> > 
> > Thank you for yout help.
> start with EM, if necessary with slightly larger box.
> > 
> > Sincerely,
> > 
> > --
> > Jordi Camps Puchades
> > Instituto Nacional de Bioinformatica (INB) Nodo 
> Computacional GNHC-2 
> > UPC-CIRI
> > c/. Jordi Girona 1-3              
> > Modul C6-E201                     Tel.  : 934 011 650
> > E-08034 Barcelona                 Fax   : 934 017 014
> > Catalunya (Spain)                 e-mail: jcamps at lsi.upc.edu
> > 
> > -----Mensaje original-----
> > De: gmx-users-bounces at gromacs.org 
> > [mailto:gmx-users-bounces at gromacs.org] En nombre de Anton Feenstra 
> > Enviado el: jueves, 19 de mayo de 2005 7:45
> > Para: Discussion list for GROMACS users
> > Asunto: Re: [gmx-users] Excessive compute time using PME
> > 
> > 
> > Jordi Camps wrote:
> > 
> > > Hello list!
> > > 
> > > Yes, you were right on the problem. I computed bad box 
> vectors which 
> > > makes the box extra-large. Now I reduced the box to it's 
> real size, 
> > > but now I get some kind of error. GXM writes an step-1.pdb and a 
> > > step0.pdb files (the second one shows that all the molecules 
> > > exploded) and core dumps. I'm working to find the cause.
> > 
> > Loook at your periodic boundaries. If you changed the box size, you 
> > may well have (bad) overlap at the 'edges'.
> > 
> > 
> --
> David.
> ______________________________________________________________
> __________
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics 
> group, Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,          75124 Uppsala, Sweden
> phone:  46 18 471 4205          fax: 46 18 511 755
> spoel at xray.bmc.uu.se    spoel at gromacs.org   
> http://xray.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> ++++++++++
> 
> 
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
>