Re: [Visualpython-users] points demo

[Anton S. <br...@po...>]

Bruce Sherwood wrote:
 > However, be aware that the star positions aren't real, just
 > generated from random numbers (and there's an odd effect in
 > the random number generator, because a strong concentration in
 > the form of a straight bar shows up at certain angles of view).

The fault in the stars lies not in the RNG.  Uniform distribution of 
latitude gives a concentration toward the z axis.  The code should be:

	theta = uniform(0,2*pi)
	zn = uniform(-1,1)	# iota = acos(zn)
	rn = sqrt(1-zn**2)
	rho = uniform(rmn*a,rmx*a)
	rxy = rho*rn
	x = rxy*cos(theta)
	y = rxy*sin(theta)
	z = rho*zn

Another way is
	x = normalvariate(mean,sigma)
	y = normalvariate(mean,sigma)
	z = normalvariate(mean,sigma)
This has spherical symmetry but may not give
the radial distribution you want.

-- 
Anton Sherwood *\\* www.ogre.nu
"How'd ya like to climb this high without no mountain?" --Porky Pine