If I get it right, it seems to me that the base of the problem is that 
POVRay uses indirect coordinate system (left hand), but it is not enough 
to take z as the up direction, you should take -z. In a direct 
coordinate system (u, v, w), the w vector should see u->v in the 
trigonometric way (which is counter clock). Putting "-" in the right 
places in your formulas should do it.

JC

Ed Jackson wrote:
> Hi all,
> 
> I was just messing around with media densities using various probability
> densities for the electron in a hydrogen atom.  These things are usually
> done in radial coordinates because the math is simpler.
> 
> The conversion from cartesian to radial coordiates is simple enough, and
> functions.inc even includes f_r, f_th, and f_ph to help--which leads me to
> my question.
> 
> From my experience in physics, it is conventional to define theta as the
> angle from the vertical axis (latitude, ranging from zero at the north
> pole to pi at the south pole), and phi as the longitudinal angle (ranging
> from zero to 2*pi).
> 
> POV, it seems, defines these in the opposite manner.  I was hoping that
> someone in the know could tell me whether this is a mistake, or if the
> convention is simply less well-defined than I believed.  After all, we
> usually use z-up cartesian coordinates in physics, too.  :)
> 
> Anyway, I'm just curious...if it really bothered me, I'd just define my
> own conversion functions.
> 
> 	Thanks,
> 	 Ed

Post a reply to this message