Hi Jean,

Jean Montambeault wrote:
>     I'd follow Alan's advice to stick to simple normal orientations,
> for the least, but I find it even easier to use only normal = y and
> then transform by rotating, translating, etc.

First I thought of a texture as some two dimensional thing being wrapped
around an object. With this picture in mind it made no no sense to me to
rotate a texture with an object (because I wouldn't have much influence
on the positioning anyway).

Now as I got some explanations I see that a texture is something like a
"space coloring rule" intersecting an object. What I see in the output
is the surface of the through and through colored object.
In this case it makes sense for me to "attach" a texture to an object.

> This way I can link the
> object (any object, not just a plane) to its texture in a declaration
> and do without fussing with the normal (just try to find the right
> rotation to align the pigment on, let's say, normal <.4, .55, 22.2> ;}
> I don't want to spend my brain muscle on that kind of exercice).

Ok, got it. I have to put my texture onto (into?) my object BEFORE
rotating and scaling it because this is the only way to guarantee the
texture's correct orientation relative to the object. When I rotate both
in common the patches/stripes on the object don't change their position.

Correct?

Thanks, Guido.
-- 
>- Live long and prosper. -<>- Guido Heer, Basel, Switzerland -<
>- http://pobox.com/~heer -<>- Home of the german Newton FAQs -<
>- Get my public PGP key @ http://pobox.com/~heer/pgpkey.html -<

Post a reply to this message