clipka <ano### [at] anonymousorg> wrote:
> Am 10.02.2017 um 11:15 schrieb omniverse:
>
> > This had me looking at slope pattern in the documentation and something I wasn't
> > aware of is the "new to 3.7" point_at for it. Or if I did know I had forgotten
> > it.
> ...
> > Food for thought anyhow, because it seems like it would be what SunPosition is
> > made for.
>
> Only if the sun is supposed to be very close, so that its rays are
> supposed to be noticeably non-parallel. Otherwise the direction vector
> would probably be the better choice.

Hmm. I think I know what you're saying, which is direction vector easier and
that point_at is intended for more localized behavior of slope.
At least that's my guess.

I just tried something with point_at anyway, going by your saying 'sun very
close'.

Took the SunPosition numbers given by it's dot x y and z.
Divided those (less decimals) into the slope SunPosition for the point_at and it
resembled the direction vector method.
Not the same as direction only, but similar perhaps.

#declare pigment3 =
pigment {
  slope {
    point_at
    <SunPosition.x/818827576,SunPosition.y/163128554,SunPosition.z/550373032>/2
....

Post a reply to this message