VMD-L Mailing List

From: Ondrej Marsalek (ondrej.marsalek_at_gmail.com)
Date: Wed Jun 10 2009 - 07:54:05 CDT

Hi,

for what it's worth, I have done some python scripting to enable the
animation of volumetric data. It uses a VMD molecule for each frame of
volumetric data and is faster than I'd originally expected. Data get
removed from memory if needed. The code is very messy at the moment,
so it is not publicly available, but I could clean it up and release,
if people are interested.

John, am I reading it right that direct support for volumetric data
animation (and "unloading") is not coming any time soon? If it was
close, there would be no point in working on my current code.

Best,
Ondrej

On Wed, Jun 10, 2009 at 14:24, John Stone<johns_at_ks.uiuc.edu> wrote:
>
> Rob,
>  At present, you can only load a volumetric data object for all
> timesteps.  For a subsequent version of VMD I may enable loading
> these for a trajectory, but to make that feasible, several parts
> of the VMD volume rendering code need to be GPU-accelerated so that
> animation of such datasets becomes usable.  For now, you
> could start with a plugin that loads only volumetric maps from the
> last timestep.  You might also consider talking to Axel and Jan
> about projecting onto GTOs and providing the QM data to VMD via
> the QM plugin APIs.  If you could do that, it would use far less memory
> than storing the volumetric grids for the properties of interest.
> We expect to add several new QM related representations in the
> next version that would likely be of interest to you.
>
> Cheers,
>  John Stone
>  vmd_at_ks.uiuc.edu
>
> On Wed, Jun 10, 2009 at 01:11:34AM -0700, Rob wrote:
>>
>>
>> Hi,
>>
>> I'm writing a plugin for Abinit.
>>
>> I can read sequential structure geometry files, so that I can
>> see the atoms move during a structure relaxation, for example.
>>
>> Each relaxation step also comes with a charge density distribution
>> as volumetric data; can I read this also as frames/timesteps?
>>
>> I'm writing the plugin right now and I see that the "read_next_timestep"
>> function is repeatedly called until it returns MOLFILE_EOF.
>> However, the "read_volumetric_data" is called only once during this
>> procedure.
>> Hence, I can have a sequence of atom positions as frames, but only
>> one volumetric data......
>>
>> Or am I doing something wrong?
>>
>> Thanks,
>> Rob.
>>
>>
>>
>
> --
> NIH Resource for Macromolecular Modeling and Bioinformatics
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
>  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078
>