sugtech@lll-lcc.UUCP (SUG Tech Committee) (08/20/86)
I have some questions concerning ray tracing and I would appreciate
anybody's help.
1.) What do you do when a ray intersects a polygon straightway (the
ray lies in the plane of the polygon)? One could have a scene in
which several polygons resided in the same plane, each one behind
the other. One would have to determine the closest one, in order to
dsisplay the proper color.
2.) What is a fast method of finding the intersection of two 3D lines?
Determinant?
3.) Are there any new ways of antialiasing other than super sampling or
stochastic sampling?
4.) Is the Cook-Torrance shading model the most accurate one to date or
is there a better one?
5.) What is the advantage of applying modelling transformations to rays
and not to the objects (see Roth's paper, 1982)? It appears that this
would slow down the ray tracing process, since the transformations need
to be applied within the double for loop. A detailed explanation
and an example would really help here.
6.) What should one do to calculate the depth of a polygon at a certain
point if the plane equation coefficent "C" is equal to zero?
Examples would be N = (0, -1, 0, 1) or N = (1, 0, 0, -1). The
forumula z = (-d -ax - by)/c does not work in this case.
7.) What are some of the visual effects that ray tracing cannot yet model?
8.) I read in the book "Procedural Elements for Computer Graphics",
p. 379, that small objects may be lost if they are viewed indirectly by a
curved surface. Is there any way to prevent this?
9.) Are there text books out now that have lengthly, detailed sections on
ray tracing?
10.) This question is concerned with the method of antialiasing used by
Whitted. Michael Sweeney, in his master thesis, stated "that the four
corners of a pixel are considered to be the same color if the sum
of the differences in their red, green and blue components is less
than this threshold (the threshold being a value specified by the
user)". Could someone please show me how the sum referred to by Sweeney
is derived, given his explanation?skinner@saber.UUCP (Robert Skinner) (08/25/86)
> > 10.) This question is concerned with the method of antialiasing used by > Whitted. Michael Sweeney, in his master thesis, stated "that the four > corners of a pixel are considered to be the same color if the sum > of the differences in their red, green and blue components is less > than this threshold (the threshold being a value specified by the > user)". Could someone please show me how the sum referred to by Sweeney > is derived, given his explanation? I have seen some of Sweeney's code, the threshold defaults to 40.0 (out of 3*256), but can be specified as a program option. ------------------------------------------------------------------------------ "You know something people? I'm not black, But there's a whole lots of times I wish I could say I'm not white" Name: Robert Skinner Snail: Saber Technology, 2381 Bering Drive, San Jose, California 95131 AT&T: (408) 435-8600 UUCP: ...{decvax,ucbvax}!decwrl!saber!skinner ...{amd,ihnp4,ittvax}!saber!skinner