From: John Stone (johns_at_ks.uiuc.edu)
Date: Mon Nov 20 2006 - 15:26:33 CST

Hi,
  I'm well into building early test versions of what will become
VMD 1.8.6, and some of the key improvements in the next version
include improved performance and memory efficiency when working
with very large structures (multiple millions of atoms). I would
be interested in getting feedback from people working with
very large structures.

The load times for very large structures have been significantly reduced
through improved efficiency and increased used of multithreading and
parallel processing.
The current test versions use just over half as much memory per-atom
compared with past versions of VMD. This allows a machine with
21GB of memory to load a single structure of over 72 million atoms,
or to load the entire PDB at once, for example.
These can be particularly helpful for large batch mode analysis jobs.
This structure memory also savings means that the same machine
running VMD 1.8.6 will have more memory available to load
trajectory timesteps than VMD 1.8.5 and earlier versions did.

The new test version also does better with some nucleic acid structures
that gave trouble with VMD 1.8.5.

The test version also contains a new "field lines" representation which
can be used to display electrostatic potential field lines, or other
flow/gradient fields from volumetric data sets.

If you're interested in trying out the latest test version which contains
these improvements, the test versions will continue to be made available
in the VMD Public Project on BioCoRE, see the access instructions here:
  http://www.ks.uiuc.edu/Research/vmd/alpha/

Your feedback and suggestions are welcome.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

-- 
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