VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Tue Nov 11 2003 - 15:36:04 CST

Hi Marc,
  No changes have been made to VMD that would in any way affect
your stereo sync. These problems are unique to the NVidia boards,
and don't occur on other hardware. They have been witnessed with other
programs besides VMD, and NVidia has acknowledged the existence of these
problems, particularly in the Quadro boards prior the FX series, if I
recall correctly. I know a few users have seen this same behavior with
multiple stereo-enabled software, so it would seem that its not in any
way specific to VMD, though the only cards I've heard of these problems
with are the NVidia boards, though I realize that's not very helpful to you.

It should not be possible for VMD to cause stereo sync to swap,
because there are two branches of code, one for the left eye, and
one for the right eye. Each branch of code sets up the viewing frustum
and the active OpenGL buffer (back left, or back right) at the same time,
so the code that sets up the viewing parameters and the code that selects
the active buffer are inseparable.

The only thing that's special about CPK or VDW are that they take by far
the most processing from the video board as they generate the highest triangle
count of any geometry produced by VMD, other than large isosurfaces....

Hope that helps clarify how all of this works. Let me know if I can be
of further assistance in narrowing this down. Have you tried any other
stereo-enabled software on your system?

Thanks,
  John

On Tue, Nov 11, 2003 at 09:13:49PM +0100, Marc Baaden wrote:
>
> Concerns: problems with stereo via Nuvision Goggles + Nvidia card
>
> Hi,
>
> having continuously had to struggle with hardware stereo problems
> (as have been mentioned by others on this list before .. basically
> the sync signal gets inverted from time to time, like changing the
> little switch on the back of the Nuvision box), I got a step further
> and have pinned down a little where these problems come from.
>
>
> It's purely phenomenological but
> - with the latest 1.8.2 beta it seems a bit better than before
> (could it be related to changes in threading code if there were any?)
> - the sync problem seems only to occur when one uses the CPK
> or licorice representation. In all other cases it is smooth
>
> Now the question is, whether CPK/licorice drawing from VMD is
> the problem, or whether these are just the ones that trigger a
> bug in eg the nvidia driver or maybe a weakness in the graphics
> board.
>
> Any comments and hints are welcome.
>
> Cheers,
> Marc
>
>
>
>
> --
> Dr. Marc Baaden - Institut de Biologie Physico-Chimique, Paris
> mailto:baaden_at_smplinux.de - http://www.marc-baaden.de
> FAX: +49 697912 39550 - Tel: +33 15841 5176 ou +33 609 843217
> 

-- 
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