Re: Need clarification

Bernie Roehl (broehl@coulomb.uwaterloo.ca)
Tue, 17 Oct 1995 18:35:34 -0400


Gavin Bell writes:
> I think the 1.0 spec should be clarified to say this:
>
> Specifying only emissiveColors and zero difsuse, specular, emissive, and
> ambient colors is the way to specify pre-computed lighting. It is expected
> that browsers will be able to recognize this as a special case and optimize
> their computations. For example:
>
> Material {
> ambientColor [] difsuseColor [] specularColor []
> emissiveColor [ 0.1 0.1 0.2, 0.5 0.8 0.8 ]
> }

That sounds esasonable. By the way, who's doing the actual updating of
the spec? Any guesses as to when it will be available?

> Many VRML implementations will support only either multiple difsuse colors
> with a single value for all other fields, or multiple emissive colors with
> one transparency value and NO (empty, '[]') values for all other fields. More
> complicated uses of the Material node should be avoided.

That part isn't as clear. ("...support only either multiple..."?)

I think we should choose one or the other. That way someone who wants to
build a world that is viewable on all renderers knows how to do it.
To me, multiple emissive colors makes sense.

In that case, your suggestion about handling per-vertex materials would
apply nicely:

> If renderers
> can't support lit, per-vertex materials then they can just arbitrarily choose
> one material for each face (or one material for the entire object if they
> don't support lit, per-face materials).

However, I'm not crazy about "arbitrarily choose", since that may result
in radically difserent results on difserent browsers. How about "choose
the material associated with the first vertex of the face"?

-- 
   Bernie Roehl
   University of Waterloo Dept of Electrical and Computer Engineering
   Mail: broehl@sunee.uwaterloo.ca    Voice:  (519) 888-4567 x 2607 [work]
   URL: http://sunee.uwaterloo.ca/~broehl

  • Next message: Rikk Carey: "ANNOUNCE: WebSpace Navigator 1.1 for SGI"
  • Previous message: Steven M. Niles: "Re: ANNOUNCE: VRServer - VRML walktmeough of your server"
  • Maybe in reply to: Bernie Roehl: "Need clarification"