From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Nov 02 2011 - 14:12:35 CDT

Bogdan,
  Yeah, I was expecting you to say that you would find that it was
the dynamic bond determination that takes the runtime and not the
graphics... :) I have been working on various new code that uses
the GPU to do the "heavy lifting" for this kind of calculation,
and in a later version of VMD it should be possible for the dynamic
bonds representations to run 10x faster using a GPU computing approach,
and possibly even faster than that.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Wed, Nov 02, 2011 at 07:42:01PM +0100, Bogdan Costescu wrote:
> Dear Axel and John,
>
> Thank you for your suggestions. I have tried already Axel's and I
> didn't see much improvement in speed, but after trying John's I have
> understood that the speed limitation is coming from the cutoff
> calculations and not from the graphical representation. I'll combine
> this with a selection such that only the interesting regions are shown
> with dynamics bond determination while the rest uses normal Lines -
> this should make it quite a lot faster; it's nice that one could set
> this per representation per molecule. Even though the dynamics bond
> cutoff is not exposed in the GUI, it's fine for me: I have asked for a
> quick-and-dirty way and got exactly that :-)
>
> Thanks again!
> Bogdan
>
> On Mon, Oct 31, 2011 at 22:18, John Stone <johns_at_ks.uiuc.edu> wrote:
> > Hi,
> >  Actually there is an undocumented way of causing the lines representation
> > to use a dynamic bond search, by setting the representation from a Tcl
> > command, adding an extra parameter to set the cutoff distance for the
> > bond search.  I have been meaning to add this to the GUI, but it will
> > require significant changes to graphical representations window, so
> > I have been waiting until several new features are exposed in the GUI
> > at the same time ... but I digress.  To enable dynamic bond determination
> > for the lines rep (and several others), just add the cutoff distance
> > as an extra representation parameter.  For the lines rep you would normally
> > have a representation command similar to this:
> >  mol modstyle 0 0 Lines 1
> >
> > To cause it to use dynamic bond determination, add the cutoff, e.g. 1.6A
> >  mol modstyle 0 0 Lines 1 1.6
> >
> > Cheers,
> >  John Stone
> >  vmd_at_ks.uiuc.edu
> >
> > On Mon, Oct 31, 2011 at 03:32:47PM -0400, Axel Kohlmeyer wrote:
> >> On Mon, Oct 31, 2011 at 2:21 PM, Bogdan Costescu <bcostescu_at_gmail.com> wrote:
> >> > Dear vmd-l,
> >> >
> >> > I would like to have some kind of DynamicBonds representation, but
> >> > where the bonds are not represented as cylinders but as Lines. I like
> >> > how DynamicBonds works, but the representation is slow because there
> >> > are many cylinders to be drawn. Is there a quick-and-dirty way of
> >> > combining the two ? I could, of course, just draw the lines myself
> >>
> >> no. at least not without changing source code
> >> and building your own version of VMD.
> >>
> >> > from TCL for each frame, but again due to speed reasons I would prefer
> >> > to use a built-in mechanism.
> >>
> >> you can reduce the cylinder resolution to 2 or 3 and diameter to 0.1
> >> depending on your graphics hardware, turning GLSL on or off might
> >> speed up things, too.
> >>
> >> axel.
> >>
> >> > Thanks in advance for any suggestion!
> >> > Bogdan
> >> >
> >>
> >>
> >>
> >> --
> >> Dr. Axel Kohlmeyer
> >> akohlmey_at_gmail.com  http://goo.gl/1wk0
> >>
> >> College of Science and Technology
> >> Temple University, Philadelphia PA, USA.
> >
> > --
> > NIH Resource for Macromolecular Modeling and Bioinformatics
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> > http://www.ks.uiuc.edu/~johns/           Phone: 217-244-3349
> > http://www.ks.uiuc.edu/Research/vmd/       Fax: 217-244-6078
> >

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
http://www.ks.uiuc.edu/~johns/           Phone: 217-244-3349
http://www.ks.uiuc.edu/Research/vmd/       Fax: 217-244-6078