[PD] The Party at the Center of the Universe

Mathieu Bouchard matju at artengine.ca
Sat Dec 10 11:29:19 CET 2005


On Sun, 4 Dec 2005, megalegoland at yahoo.fr wrote:

>  Well I am working on PMPD and particulary on the morphing exemple for a
>  live show in two weeks, my computer (athlon 2600+, 333Mhz bus freq, 512
>  RAM, and Nvidia 64Mo) can handle the 125 spheres with no problems, it's
>  slowing down if I use text with 'round' fonts and more than two
>  letters, So I could make an emulation of the largest planets (from
>  mercury to pluto-charon) with our sun without problems, but the solar
>  system has more than four hundred thousands planets

Well, please make a distinction between the computation time and the 
rendering time. Also please take into account that it's pointless to try 
to display all of them at once, and maybe even 1% of them.

>  So, a cluster of several computers (maybe with pure-dyne) would be
>  needed to make an accurate model, indeed...

I would reach that conclusion but rather based on the fact that computing
gravity is an O(n^2) process, due to the N*(N-1)/2 gravitational
interactions that I was talking about. So I wouldn't even take rendering
into account here.

Anyway, any rendering I'd do would only involve some objects, because
humans can't really track 400000 objects on screen, even with the
biggest-ass monitors possible (normal screens don't have nearly enough
pixels).

 _ _ __ ___ _____ ________ _____________ _____________________ ...
| Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju
| Freelance Digital Arts Engineer, Montréal QC Canada




More information about the Pd-list mailing list