[Chimera-users] 3D stereo

[Greg Couch]

On 07/21/2010 07:24 AM, Nadir T. Mrabet wrote:
> Hi,
>
> This is for teaching purposes so that students can "walk" though 3D
> structures.
> Would it be possible to use Chimera via a 3D stereo beamer to project
> display on a very large screen and watch the selected structure in 3D by
> means of shutter glasses?
> If this is so, what kind of equipment would you recommend?
>
> Many thanks,
>
> Nadir
>    

The short answer is yes, it is possible.  For longer answers see recent 
postings on the chimera users list: 
<http://plato.cgl.ucsf.edu/pipermail/chimera-users/2010-May/005145.html>, and 
<http://plato.cgl.ucsf.edu/pipermail/chimera-users/2010-June/005297.html>.  
There will probably be some announcements of new stereo hardware at the 
SIGGRAPH conference next week, so the choices may change soon.

And here's some generic advice about choosing a graphics card:

The answer to your questions depends on what kind and what amount of 
data you're planning to view with chimera.  For instance, if you're 
looking at volume data, large molecules, or trajectories, a graphics 
card with more memory would be important.  And if you're looking at 
large molecules in  ball and stick mode, doing volume calculations, or 
making movies, you'll want a fast CPU as well.  And if you want the best 
3D graphics in stereo, you'll need to get a workstation graphics card 
with a stereo DIN connector.

So take a look at the chimera benchmarks page, 
http://plato.cgl.ucsf.edu/trac/chimera/wiki/benchmarks, to get an idea 
of the relative performance of graphics cards with chimera.  You will 
see that we need more benchmarks from recent graphics cards (so whatever 
you get, please submit the chimera benchmarks for it).  You will also 
see that the consumer graphics cards do very well with the Volume 
depictions.  And that the workstations graphics cards generally do 
better with the Molecule scores, and that the Molecule scores are more 
dependent on the CPU speed than the Volume scores.  You will also see 
that Linux systems tend to benchmark higher than Windows systems with 
the same hardware.

     Good luck,

     Greg