ABSTRACT
In this chapter we will discuss the major computer representation of the real numbers, floating-point, along with the associated bitwise formats, basic operations, features, and limitations. By design, we will transition from general mathematical discussions of number representation toward implementation-related topics of specific relevance to 3D graphics programmers. Most of the chapter will be spent on the ubiquitous Institute of Electrical and Electronic Engineers (IEEE) floating-point numbers, especially discussions of floating-point limitations that often cause issues in
3D pipelines. A brief case study of floating-point-related performance issues in a real application is also presented.
