LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Commands

write_dump command

Syntax:

write_dump ID group-ID style N file dump-args modify dump-modify-args 

Examples:

write_dump 1 all atom 0 dump.atom
write_dump 1 subgroup atom 0 dump.run.bin
write_dump 1 all custom 0 dump.myforce.* id type x y vx fx
write_dump 1 flow custom 0 dump.%.myforce id type c_myF[3] v_ke modify sort id
write_dump 1 all xyz 0 system.xyz modify sort id elements O H 
write_dump 1 all image 0 snap*.jpg type type size 960 960 modify backcolor white
write_dump 1 all image 0 snap*.jpg element element &
   bond atom 0.3 shiny 0.1 ssao yes 6345 0.2 size 1600 1600  &
   modify backcolor white element C C O H N C C C O H H S O H 

Description:

Dump a single snapshot of atom quantities to one or more files for the current state of the system. This is a one-time operation, in contrast to the dump command which will write out multiple snapshots periodically during a running simulation.

The syntax for this command is identical to that of the dump and dump_modify commands as if they were concatenated together, with one exception. This is so that the full range of dump_modify options can be specified for the single snapshot, just as they can be for multiple snapshots. The syntax exception is that if a second set of dump-modify-args is specified, they must be preceeded by the "modify" keyword, so that LAMMPS can cleanly separate the two sets of args.

Note that the dump ID and frequency N arguments are ignored, since they are irrelevant when only a single snapshot is being written.

Also note that if the specified filename uses the wildcard characters "*" or "%", as supported by the dump commmand, they will operate in the same fashion to create the new filename(s).


Restrictions:

All restrictions for the dump and dump_modify commands apply to this command as well. For the dump image dump style, this means the filename is required to contain a '*' character, which will be replaced by the current timestep.

Since dumps are normally written during a run or energy minimization, the simulation has to be ready to run before this command can be used. Similarly, if the dump requires information from a compute, fix, or variable, the information needs to have been calculated for the current timestep (e.g. by a prior run), else LAMMPS will generate an error message.

For example, it is not possible to dump per-atom energy with this command before a run has been performed, since no energies and forces have yet been calculated. See the variable doc page sectinn on Variable Accuracy for more information on this topic.

Related commands:

dump, dump image, dump_modify

Default:

The defaults are listed on the doc pages for the dump and dump image and dump_modify commands.