I doubt that the trig-version gives a good distribution: inlineVectortrig(uniform_01<generator_type>&random) { constfloatphi=2*M_PI*random(); constfloatcos_theta=2*random()-1; constfloatsin_theta=sqrt(1-cos_theta*cos_theta); returnVector(cos_theta,sin_theta*cos(phi),sin_theta *sin(phi)); } The result ought to be more dense at the poles. /$ 2007/6/5, more effective thinking in the exceptional C++ programming language <effective.thinking@gmail.com>:
Dear,
I'm a newcomer on the boost list. I subscribed to share the results of my experiments on picking random points on sphere. This procedure is often a bottleneck for very fast Monte-Carlo simulations of physical process where you must pick vectors in random directions. I compiled a document that explain the experiments and the results. The conclusion contains some propositions to modify boost::random_on_sphere and I would like to know if a new implementation following the presented ideas would be welcome.
Colas Schretter
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost