Path: utzoo!attcan!uunet!lll-winken!lll-lcc!ames!pasteur!agate!saturn!ucscb.UCSC.EDU!easterb
From: easterb@ucscb.UCSC.EDU (William K. Karwin)
Newsgroups: comp.graphics
Subject: Re: Raytracing speedups....
Summary: good for raytracing previewer
Message-ID: <5796@saturn.ucsc.edu>
Date: 16 Dec 88 06:42:13 GMT
References: <907@cmx.npac.syr.edu>
Sender: usenet@saturn.ucsc.edu
Reply-To: easterb@ucscb.UCSC.EDU (William K. Karwin)
Organization: Uncle Charlie's Summer Camp
Lines: 36
Keywords:

In article <907@cmx.npac.syr.edu> gefuchs@top.cis.syr.edu (Gill E. Fuchs) writes:
>koussari@rodan.acs.syr.edu         gefuchs@top.cis.syr.edu
>...transform the object center plus any point
>along the border (radius for sphere) to screen coordinates.
>    For each all objects create an array which contains coordinates of
>opposite sides of a rectangle that bound that object...
>... you check to see if the current screen(x,y) coordinate that
>is being processed is in side any of the bounding boxes.
>   If it is not in anyof the boxes, then get the background color.
>   If it is then intersect the ray with the spheres pointed to by the
>structure that holds what objets that exist inside that bounding box.

The way I am reading this, you are generating 2d boxes in screen
coordinates, not 3d boxes around the objects in world coordinates.

>   Would there be problems with secondary rays? (reflected,refracted)

Yes, I think there would be some problems with secondary rays.
Your purpose for these 2d bounding boxes is to make a lot of the
intersection computation work overhead, to be done once at the
beginning of the job.  Casting a secondary ray would be equivalent
to changing the viewpoint of the scene, and so would require another
entire set of 2d bounding boxes to be created.  This would have
to happen during runtime, because you could not predict secondary rays.

I have seen the results of ray casting algorithms that only display
regions of sharp contrast changes, i.e., edges of displayed objects.
These algorithms do no reflection, refraction, or shading.  At best they
do constant-shaded coloring of surfaces.  They are most often used (to my
knowledge) for fast generation of images in a CAD application, or as
a scene previewer for a more realistic renderer.

I think your idea would be very well suited for such a display method.

William K. Karwin                 ARPA  : easterb@ucscb.ucsc.EDU
"Any nitwit can understand        UUCP  : ...!ucbvax!ucscc!ucscb!easterb
computers.  Many do." -T. Nelson  BITNET: easterb@ucscb.BITNET