From: Axel Kohlmeyer (
Date: Mon Aug 29 2011 - 17:26:22 CDT

On Mon, Aug 29, 2011 at 5:35 PM, Chris Knorowski <> wrote:
> Is it possible to load a trajectory where the number of particles in each
> frame aren't conserved?
> I initially tried to load a trajectory in xyz format but vmd only loaded the
> first frame, I figured this was because the second had a different number of
> particles.

nod. that is how VMD is designed.

> Since this didn't work I tried something else as well. I kept the same
> number of particles, but changed the names of the ones that should
> be disappearing,
> for example:
> the first frame has 5 "Z" particles and 10 "M" particles. The second frame
> has 7 'Z' particles and 13 "M" particles. And then I wanted to show only the
> Z particles in each timestep.
>  However, it seems that vmd uses the names of the particles in the first
> frame and doesn't reupdate the name list as I step through time.

nod. the names of all subsequent frames are completely ignored
and VMD assumes that particles remain the same over the course
of a trajectory.

> Is there a way to force vmd to re-update the names at each timestep. Or make
> vmd load a trajectory with varying numbers of particles in each timestep.

this is a recurring question. this is due to a fundamental design decision,
which optimizes VMD performance and reduces its memory footprint for
the most common use cases. you can work around this with scripting.
one option to do this has been implemented in the topotools plugin:

for other suggestions, please search the mailing list archives.


> Thanks,
> Chris
> --
> Christopher Knorowski
> Iowa State Physics Department

Dr. Axel Kohlmeyer
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.