Path: utzoo!attcan!uunet!wuarchive!rex!samsung!cs.utexas.edu!helios!gordon
From: gordon@cs.tamu.edu (Dan Gordon)
Newsgroups: comp.graphics
Subject: Re: Graphics Texts etc.
Message-ID: <8256@helios.TAMU.EDU>
Date: 14 Sep 90 16:53:44 GMT
References: <101880023@hpcvlx.cv.hp.com> <23900016@sunc7>
Sender: usenet@helios.TAMU.EDU
Organization: Computer Science Department, Texas A&M University
Lines: 20

In article <23900016@sunc7> mcooper@sunc7.cs.uiuc.edu writes:
<
<There is also "Computer Graphics" by Hearn and Baker.
<
<Not as thurough as Foley, but more suited, IMHO, to a single semester course.
<It's sort of an overview of most of the basic graphics topics.
<
<But then, since Hearn was my prof for two years, I might be biased..

Maybe you can clear up something that has been a mystery to me: why
does Hearn & Baker use a most inefficient polygon scan conversion 
algorithm? Is it to avoid using linked lists? If you look up polygon
scan-conversion in Foley & van Dam (old and new versions) and in many
other books, you will see that a simple linked list is used to maintain
the active edge table (or list). 

Polygon scan-conversion is perhaps the single most important algorithm
in computer graphics. It can be extended in so many ways to yield
hidden surface removal, Gouraud shading, and so on. Its efficient
implementation is of the utmost importance.