Tom Melly <tom### [at] tomandlucouk> wrote:
> A plane stretching to infinity would, presumably, need an infinite bounding box
> (otherwise it's not infinite).

  I think an infinite object does not have a bounding box at all.

> An infinite number of rays would be required to
> trace it, since there would be an infinite number of intersection points.

  I don't understand how you came to this conclusion.
  An infinite number of intersection points to calculate what? For one
pixel we need one intersection point (unless using antialiasing, which
means we may need something like 9 intersection points). What do you need
to calculate an infinite amount of intersection points for?

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message

"Warp" <war### [at] tagpovrayorg> wrote in message news:3e8175ca@news.povray.org...
>
>   I don't understand how you came to this conclusion.

You obviously forgot to factor in my idiocy.

>   An infinite number of intersection points to calculate what? For one
> pixel we need one intersection point (unless using antialiasing, which
> means we may need something like 9 intersection points). What do you need
> to calculate an infinite amount of intersection points for?

This is the solution to my confusion (and fairly bleedin' obvious when I think
about it). I was thinking that the number of rays required to trace an object
was dependant on the object's actual size (infinite in the case of a plane),
whereas the number of rays is more dependant on the area of the image that the
object occupies (always finite).

Post a reply to this message

"Warp" <war### [at] tagpovrayorg> wrote:
>
>   I think an infinite object does not have a bounding box at all.
>
Internally, infinite objects are even kept in a separate list, at scene
root, outside of any hierarchies.

Post a reply to this message