[Beowulf] 3d rendering cluster

[Robert G. Brown]

On Tue, 24 May 2005, Paul K Egell-Johnsen wrote:

> > Dual cores are of course de rigeur at the moment, dahlings :-)
> > We are getting good results with them.
> > 
> > I would be interested to see some real-world findings with 3D rendering
> > on dual core boxes. It would be worth doing some research, and getting
> > some benchmarking time before making a commitment. If you do make any
> > comparisons, please come back and tell us.
> 
> As per suggestion I've turned this question over to the software
> vendor, they aren't into selling boxes, therefore I believe any
> numbers they can serve up would be of interest.
> 
> I also asked them for numbers on what L2 cache might do for the
> rendering speed, as well as memory size and disk cache speeds. These
> would of course have different impact on different sizes of
> renderings, scene geometry, textures and passes the rendering engine
> needs to do on each pixel.

Trust only benchmarks you run yourself, in context.  Vendors can be
surprisingly ignorant of what they sell quite independent of their
possible vested interests.

There are exceptions to this, of course.  Some very trustworthy and
smart vendor reps are on this list.  But you can only LEARN to trust
them by doing your own benchmarks and tests, or asking around...

   rgb

> 
> I'm also turning to the specialized video mailing list and some 3d
> mailing lists.
> 
> Best regards,
> Paul
> 
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> 

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu