Chapter 35: Floating point arithmetic
Abstract
In the physical sciences, we usually work with real numbers, so it’s convenient to pretend that computers can do this too. However, since numbers in a computer have only a finite number of bits, most real numbers cannot be represented exactly. In fact, even many fractions cannot be represented exactly, since they repeat; for instance, \(1/3 = 0.333 \ldots\), which is not representable in either decimal or binary.
Fields of Study & Subjects
Numerical Methods and Algorithms
Mathematical and Computational Methods
Mathematical functions and polynomials
Publication details
- Journal / source
- Society for Industrial and Applied Mathematics eBooks
- Pages
- 337–349
- Publication date
- 19 January 2026
- Language
- EN
- Document type
- Book chapter