[gmx-users] maximum number of molecules

[Rainer Boeckmann]

Hi,

a few weeks ago there was a question concerning the maximum number of atoms
allowed in GROMACS (2^31-1). I now realized that the maximum number of molecules
is limited to about 100.000 - not in mdrun but one gets problems in preparing
large systems using editconf. editconf may write wrong coordinates and/or
velocities when converting files.

Rainer

--
Rainer B?ckmann
Theoretical Molecular Biophysics Group
Max-Planck-Institute for Biophysical Chemistry
Am Fa?berg 11
D-37077 G?ttingen

Tel.: [phone], Fax.: [phone]
email: rboeckm at gwdg.de

[David van der Spoel]

Post by Rainer Boeckmann
Hi,
a few weeks ago there was a question concerning the maximum number of atoms
allowed in GROMACS (2^31-1). I now realized that the maximum number of molecules
is limited to about 100.000 - not in mdrun but one gets problems in preparing
large systems using editconf. editconf may write wrong coordinates and/or
velocities when converting files.

No it works. Just try it. The largest runs in gromacs we have done here
are 450,000 atoms.

Post by Rainer Boeckmann
Rainer
--
Rainer B?ckmann
Theoretical Molecular Biophysics Group
Max-Planck-Institute for Biophysical Chemistry
Am Fa?berg 11
D-37077 G?ttingen
Tel.: [phone], Fax.: [phone]
email: rboeckm at gwdg.de
_______________________________________________
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.

[Rainer Boeckmann]

Post by David van der Spoel

Post by Rainer Boeckmann
Hi,
a few weeks ago there was a question concerning the maximum number of atoms
allowed in GROMACS (2^31-1). I now realized that the maximum number of molecules
is limited to about 100.000 - not in mdrun but one gets problems in preparing
large systems using editconf. editconf may write wrong coordinates and/or
velocities when converting files.

No it works. Just try it. The largest runs in gromacs we have done here
are 450,000 atoms.

hmm...not in my case. editconf writes wrong coordinates when for example converting
a g96-file into a pdb-file. In contrast it writes correctly a g96-file from the tpr.
grompp does wrong when writing a tpr-file from a g96, correct when using a pdb-file.
I think that this depends on the number of molecules (more than 100,000 in my case),
not on the number of atoms and appears only when converting a g96-file, but I tested
not all combinations...

Rainer


--
Rainer B?ckmann
Theoretical Molecular Biophysics Group
Max-Planck-Institute for Biophysical Chemistry
Am Fa?berg 11
D-37077 G?ttingen

Tel.: [phone], Fax.: [phone]
email: rboeckm at gwdg.de

[David van der Spoel]

Post by Rainer Boeckmann

Post by David van der Spoel

Post by Rainer Boeckmann
a few weeks ago there was a question concerning the maximum number of atoms
allowed in GROMACS (2^31-1). I now realized that the maximum number of molecules
is limited to about 100.000 - not in mdrun but one gets problems in preparing
large systems using editconf. editconf may write wrong coordinates and/or
velocities when converting files.

No it works. Just try it. The largest runs in gromacs we have done here
are 450,000 atoms.

hmm...not in my case. editconf writes wrong coordinates when for example converting
a g96-file into a pdb-file. In contrast it writes correctly a g96-file from the tpr.
grompp does wrong when writing a tpr-file from a g96, correct when using a pdb-file.
I think that this depends on the number of molecules (more than 100,000 in my case),
not on the number of atoms and appears only when converting a g96-file, but I tested
not all combinations...

I see, pdb file might be a problem, but gro and g96 work.

[Berk Hess]

I think gro does not have a maximum number of molecules,
since all numbers are printed modulo 100000 (modulo 10000
for residue numbers in pdb).
But g96 and pdb do have problems, there the prints should
also be modulo 100000 (modulo 10000000 for atom numbers in
g96).
Gromacs does not use the atom numbers and residue numbers
are only used to distinguish between different residues.

Berk.




_________________________________________________________________
MSN Photos is the easiest way to share and print your photos:
http://photos.msn.com/support/worldwide.aspx

[Berk Hess]

Post by Berk Hess
I think gro does not have a maximum number of molecules,
since all numbers are printed modulo 100000 (modulo 10000
for residue numbers in pdb).
But g96 and pdb do have problems, there the prints should
also be modulo 100000 (modulo 10000000 for atom numbers in
g96).
Gromacs does not use the atom numbers and residue numbers
are only used to distinguish between different residues.

Sorry, pdb has no problems, only g96.
I fixed the g96 writing for the next release.

Berk

_________________________________________________________________
Join the world?s largest e-mail service with MSN Hotmail.
http://www.hotmail.com