VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Tue Jun 27 2006 - 17:00:26 CDT

Hi,
  Since you didn't mention what video card you currently have I can't
immediately give you a suggestion on whether or not it's really up
to this task. I have to ask though, do you _really_ want to view structures
of 300,000 atoms in full atomic detail for the entire structure? If you
do, then I can give some suggestions depending on what video card you're
using currently. An easy way to get MUCH faster rendering for such a huge
structure if you want atomic detail is to use the "Points" representation,
"Lines", and so on instead of CPK. They can be quite useful if you tweak
the point size, use depth cueing, etc.

  John Stone
  vmd_at_ks.uiuc.edu

On Tue, Jun 27, 2006 at 10:24:16AM -0400, Sacerdoti, Federico wrote:
> Hi,
>
> We are running into performance problems with VMD analysis of very large
> systems (300,000 atoms), especially when the "CPK" mode is used to
> render all atoms as spheres. The machine running VMD immediately slows
> to a crawl when asked to rotate the system. It has a 3.4GHz Xeon cpu,
> 2GB of ram, and runs Windows XP.
>
> I am tempted to immediately upgrade the video card of the machine, but I
> wonder if there are other bottlenecks. Does this list know if more CPU
> or memory would be necessary (or just as effective)? More CPU would mean
> running VMD on a multi-processor SMP, and more memory would mean on the
> order of 4-8GB.
>
> Thanks,
> Federico
>
> --
> D.E.Shaw Research LLC
> New York NY 

-- 
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