Wasn't it Ross Litscher who wrote:
>Mike Williams wrote:
>> 
>> Both problems are due to the fact that POV is inconsistent about the way
>> it uses angles. The ROTATE operation uses angles measured in degrees,
>> but its trig functions use angles measured in radians. You used degrees
>> throughout.
>> 
>> If you replace both occurrences of "cos(theta..." by "cos(theta/r_d..."
>> (you've already declared r_d to be 180/Pi) then both orbits become nice
>> smooth ellipses and the two collision points become clearly visible.
>> 
>
>
>thanks for clearing this up. is there a logical reason for the
>inconsistence?

I'd guess that the operations were written by different people. ROTATE
was always part of the POV language, and I seem to remember that it was
part of the DKB_Trace language that preceded POV_Ray.

The trig functions are comparatively recent additions. I'd guess that
the guy who added them was very familiar with the trig functions in the
C language (in which POV_Ray is currently written) which work in
radians.

-- 
Mike Williams * ##
Gentleman of Leisure

Post a reply to this message