Inside Macintosh: PowerPC Numerics

Part 1 - The PowerPC Numerics Environment

This part is a general description of PowerPC Numerics. Chapter 1 describes the standards for floating-point arithmetic that PowerPC Numerics implements (IEEE Standard 754 and the FPCE technical report) and discusses why these standards are important. If you are unfamiliar with how computers perform floating-point arithmetic, you should read Chapter 1. Chapters 2 through 6 describe how PowerPC Numerics implements the standards. They describe the basic features shared by all PowerPC Numerics implementations, including

• the numeric data formats
• the special values NaN (Not-a-Number) and Infinity
• the methods by which floating-point expressions are evaluated
• environmental controls, such as setting the rounding direction and handling exceptions
• conversions between the different numeric formats
• operations supported by PowerPC Numerics

Although Part 1 uses the C programming language in its examples, many of the facilities of PowerPC Numerics are accessible to users of virtually any high-level programming language, as well as to assembly-language programmers.

---

Part Contents

Chapter 1 - IEEE Standard Arithmetic

Chapter 2 - Floating-Point Data Formats

Chapter 3 - Expression Evaluation

Chapter 4 - Environmental Controls

Chapter 5 - Conversions

Chapter 6 - Numeric Operations and Functions