Christian Froeschlin <chr### [at] chrfrde> wrote:
> Trevor G Quayle wrote:
>
> > For any CSG, a bounding box is made as a single object that is known to contain
> > everything in the CSG.  So for an intersection operation, the bounding box will
> > be big enough to contain both (or all) your objects, regardless of how big the
> > remaining part is.
>
> Are you sure about this? The intersection of two finite
> objects is contained in the intersection of their bounding
> boxes so this should be more efficient than you describe.
> In fact it seems to work quite well for my intersections,
> even though one of the shapes is large.
>
> There is only an issue when intersections turns out
> to be empty and I think this is a fixable problem.
>
> > This is even more extreme in difference operations as POV treats them as an
> > intersection of the first object with inverses of the subtracting objects.
> > Essentially the subtractiong objects are infinite in size.
>
> Of course, an infinite box doesn't help much. Still I'd expect
> the bounding box of the difference to be the bounding box of the
> original finite object (which is still slow for many holes).

You may be right.  I am uncertain I guess how intersection is treated.  Perhaps
it is just an issue with non-intersecting boxes.

For subtraction, I do know I have tried in the past as an example to make a golf
ball from a sphere with many spheres subtracted.  This was unbearably slow.  But
switching to a blob with negative blob dimples was quite fast by comparison.

Maybe better answered by someone who knows the inner workings of bounding.

-tgq

Post a reply to this message