That is kinda what I was going for, a solution that while not perfect it
is easy to implement.  I actually kind of considered the variable angle
as a solution.  Just after I posted the message.  Let me know how it
turns out.

Steve


Margus Ramst wrote:
> 
> Stephen Lavedas wrote:
> >
> > My solution was merely to increase the angle to something like >95
> > degrees or so until results are good.  Actually, I don't think that for
> > your hair at least that 90 degrees is going to be enough...ie, since the
> > hair sticks up from the object and is visible beyond the surface.  My
> > solution would draw excess hairs, but fewer than covering the entire
> > surface
> >
> > Steve
> 
> With some objects, it would work. With others, it wouldn't. Consider a box.
> There is a very sharp change in normal direction at the edge. The
> normal/camera difference could be up to 180 degrees, yet the point could lie
> next to a visible point and culling shouldn't take place. You could argue that
> the existing hairs would obscure this shortcoming. But if the box were very
> thin, the lack of hairs on the backside would become visible.
> Still, this solution is nice and simple. Anything better would need to have
> some kind of a sorting algorithm, I'm afraid. Perhaps I'll let the user
> specify a culling angle. 180 degrees - no culling. 0 degrees - nice and bald
> :)
> 
> Margus

Post a reply to this message