Nicolas Calimet wrote:
>>Do you think recalculating a bounding tree after every simulation step
>>will be faster in the end?
> 
> 
>         I don't know how collision detection is performed in MegaPOV fo
r now: does
> it account for all particles at once ?  (i.e. O(N^2) time complexity ?)


That depends on your settings, if you do full collision detection you of 

course need to check every topology feature against every other one - O(n


>         From my experience in biomolecular simulations, there is in pri
nciple no
> need to recalculate the interaction list (or bounding tree) at every st
ep
> of the simulation.  [...]

I am not sure what you mean by 'interaction list' but this is not like a 

molecular simulation where you have very high number of molecules and 
are only interested in the statistical outcome.  A mass not correctly 
colliding where it should will ruin the whole simulation.  The idea of 
using techniques known from bounding (i.e. min/max metric to determine 
the possibility of a collision) is surely not bad but this will not 
change the order of complexity.

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 07 Mar. 2004 _____./\/^>_*_<^\/\.______

Post a reply to this message