Re: Strange problem with coordindex

Chris Marrin ([email protected])
Wed, 14 Jun 1995 12:34:13 -0700


On Jun 14, 10:15am, Ian Leicht wrote:
> Subject: Strange problem with coordindex
> I am trying to make the following shape:
> I have labeled the points with their index number.
>
>
> 4 +-----+ 5
> 2 3 | |
> +-----+ |
> 0 1| |
> +-----+ +-----+
> | | 7 6
> | +-----+
> | | 9 8
> +-----+
>
> 11 10
>
>
> If I use coordindex with 0,1,2,3,4,5,6,7,8,9,10,11,-1 the top ridges
don't
> appear. If I use 1,2,3,4,5,6,7,8,9,10,11,0,-1 the "0 1 2" ridge appears
but
> the "2 3 4" ridge i
  • ed up.
    >
    > The coordinates are correct, if I usse coordindex to make the 3 squares
    > that is okay. Why can't I trace the edge?

    By default VRML assumes all face sets to be convex. This shape is concave
    so results are unpredicatable. In webspace it fills in a couple of the
    concave edges. This can be overridden with the ShapeHints node.

    Another interesting issue is that of automatic normal generation. Since
    you do not specify normals they are being generated automatically. In the
    case of webspace they get generated with a counter clockwise orientation.
    So since you ordered your points clockwise the normals are pointing away
    from you and the back of the polygon is being lit. This may have been
    your intent but it does bring up the issue of automatic normal generation
    in VRML.

    Very little is said about normal generation in VRML 1.0. We should
    rectify this in 1.1. I'd like to suggest adding a statement that, by
    default, VRML generates counterclockwise normals.

    Opinions?

    -- 
    chris marrin      Silicon      http://www.sgi.com/Products/WebFORCE/WebSpace
    (415) 390-5367    Graphics     http://reality.sgi.com/employees/cmarrin/
    [email protected]   Inc.         
             "As a general rule, don't solve puzzles that open portals to Hell."
    

    • Next message: Keith Ahlstrom: "Worldview"
    • es"
    • Previous message:28.html">Paul Barthelemy: "Re: SPEC: DNS conventions & "naming" of VRML sites"