Targeted Molecular Dynamics(TMD): Infinite looping during read in file

LAMMPS LAMMPS Mailing List Mirror
1

Hi all,

As previously posted by Enrique Lomba at:

http://lammps.sandia.gov/threads/msg54780.html

I also encountered the same problem; the code ran into infinite loops when read in the target file. I’ve tried several versions (since 2015) but all were stuck at the point reading in target file. Has anyone successfully fixed this problem? Thanks!

2

Hi all,

As previously posted by Enrique Lomba at:

http://lammps.sandia.gov/threads/msg54780.html

I also encountered the same problem; the code ran into infinite loops when
read in the target file. I've tried several versions (since 2015) but all
were stuck at the point reading in target file. Has anyone successfully
fixed this problem? Thanks!

this almost certainly means that your target file is not correctly formatted.

axel.

3

question: which version of LAMMPS exactly are you using and how many
atoms do you have in your target file?

axel.

4

Hi Axel,

Thanks for reply. I tried version 15May2015 and 16Feb2016. The atom number used is 863 (I am just testing TMD on a vacancy diffusion problem in bulk Cu. Group id “all” is used, which corresponds to 863 atoms. The targeted configuration was obtained by averaging atomic positions of the diffused configuration at 300 K over 100 ps. The targeted file only contains 863 lines of “id x y z”, where x, y and z are unwrapped equilibrium positions of the diffused configuration).

5

Hi Axel,

Thanks for reply. I tried version 15May2015 and 16Feb2016. The atom number
used is 863 (I am just testing TMD on a vacancy diffusion problem in bulk
Cu. Group id "all" is used, which corresponds to 863 atoms. The targeted
configuration was obtained by averaging atomic positions of the diffused
configuration at 300 K over 100 ps. The targeted file only contains 863
lines of "id x y z", where x, y and z are unwrapped equilibrium positions of
the diffused configuration).

as far as i can tell, the code has seen only cosmetic changes over the
last 4 years. but one can see, that its parser for the target files is
extremely picky and unforgiving. there must not be any comments or
excess empty lines and there must be a proper EOL at the end of the
file. it also has a bug, but to trigger it, you need 997 target atoms
or more.

i've made some changes to make the parser more forgiving and fixed the
1000 lines bug and listed it for inclusion into the next LAMMPS patch.
https://github.com/lammps/lammps/pull/186

axel.

6

Many thanks, Axel. Now it works!