FYI, the dipole example script on the movies page does not work. (http://lammps.sandia.gov/movies.html#dipole) It should be replaced with the updated version given in the examples folder in the distribution.
Also, does anybody have the script to convert the output x y z mux muy muz file into a format readable by VMD as described on the movie page? i.e. “The viz is done by displaying two particles at each site where they are displaced slightly from each other in the direction of the dipole orientation.” Shouldn’t be too hard to write such at thing, but why reinvent the wheel?
Thanks,
Dave
FYI, the dipole example script on the movies page does not work.
(http://lammps.sandia.gov/movies.html#dipole) It should be replaced with the
updated version given in the examples folder in the distribution.
Also, does anybody have the script to convert the output x y z mux muy muz
file into a format readable by VMD as described on the movie page? i.e. "The
viz is done by displaying two particles at each site where they are
displaced slightly from each other in the direction of the dipole
orientation." Shouldn't be too hard to write such at thing, but why reinvent
the wheel?
dave,
wouldn't it be even more convenient, if the lammpsplugin for VMD
had some magic flag that would do that automatically for you?
i've been thinking about this for a while (primarily as a hack to represent
elliptical particles until the molfile plugin API is expanded to move arbitrary
data into VMD, which is on the TODO list but a bit complicated).
if you can provide me some example input for testing, i might be able
to have something ready rather quickly.
cheers,
axel.
Updated the script on the WWW page.
Attached is the dipole.py module I added to Pizza.py
to make the movie. It basically adds one function
that takes a dump file and a delta (separation), and turns
one atom into two (separated by delta) for each dipole
in the dump.
Then with Pizza.py you can viz it directly or write out
a new dump file, or convert it to some other format.
Steve
dipole.py (3.13 KB)
an enhanced lammpsplugin for the VMD molfile library is now available at
http://sites.google.com/site/akohlmey/software/lammps-icms/vmd-lammpsplugin-20110326.tar.gz?attredirects=0&d=1
if you set the environment variable LAMMPSDIPOLE2ATOMS to the value of a scaling
factor, e.g. 0.5, then it will read two atoms instead of one and the
two atoms will be
+/- scaling factor * dipole from the given atom coordinate. i.e.
orientation of the dipole
can be seen as well as it (relative) magnitude.
if you unset that variable, the reader will work as before.
enjoy,
axel.
Thanks Axel and Steve. I’ll check this out when I get a chance. I currently have student trying some simulations with dipoles. He’ll be grateful for your help.
Another question for Axel, how long does it take for changes to lammps to get migrated to lammps-icms? Steve posted an updated version of airebo last week. I just “git pulled” the latest lammps-icms but didn’t see an updated airebo on the list.
Thanks,
Dave
Thanks Axel and Steve. I'll check this out when I get a chance. I currently
have student trying some simulations with dipoles. He'll be grateful for
your help.
Another question for Axel, how long does it take for changes to lammps to
get migrated to lammps-icms? Steve posted an updated version of airebo last
week. I just "git pulled" the latest lammps-icms but didn't see an updated
airebo on the list.
usually, it doesn't take very long. since it is very little effort.
however, when
i have to replicate the changes in the modules that are only in my tree and
not yet in the main repo, then it can take a little longer.
i like to keep lammps-icms as close as possible to steve's repository, since
then the merging effort is small. however, his recent changes to the memory
management made me hesitate a little bit, since it always takes a few days
until people have run into the most obvious problems, so i wanted to wait that
out. also i will have to replicate those changes to a lot of the additional code
that is in lammps-icms. and even more so for the AIREBO code, since i have
an OpenMP version of that, too.
add to that, that i have been recently (overloaded) with (single
handedly) installing
and configuring a large cluster, and have a very time consuming outreach project
pending, then you can imagine that it will take a little longer. since
i need to run
benchmarks on the new machine, i have sufficient incentive to update the icms
code as well, but i just haven't found the time yet.
sorry,
axel.
Thanks Axel. No worries, of course I understand the busy factor. I just didn’t know how automated the process was. I’ll test with standard Lammps for the time being. Maybe shoot me a note if your remember when it gets x’ed of the to do list.
Also, I’ve been looking at your “interactive MD” site. I think I’ll try to get a student to tackle that as a project for the summer. It something I’ve always wanted to try.
Thanks as always for your service to the community.
Dave