The number one requirement for computer arithmetic has always been speed. It is the main force that drives the technology. With increased speed larger problems can be attempted. To gain speed, advanced processors and pro gramming languages offer, for instance, compound arithmetic operations like matmul and dotproduct. But there is another side to the computational coin - the accuracy and reliability of the computed result. Progress on this side is very important, if not essential. Compound arithmetic operations, for instance, should always deliver a correct result. The user should not be obliged to perform an error analysis every time a compound arithmetic operation, implemented by the hardware manufacturer or in the programming language, is employed. This treatise deals with computer arithmetic in a more general sense than usual. Advanced computer arithmetic extends the accuracy of the elementary floating-point operations, for instance, as defined by the IEEE arithmetic standard, to all operations in the usual product spaces of computation: the complex numbers, the real and complex intervals, and the real and complex vectors and matrices and their interval counterparts. The implementation of advanced computer arithmetic by fast hardware is examined in this book. Arithmetic units for its elementary components are described. It is shown that the requirements for speed and for reliability do not conflict with each other. Advanced computer arithmetic is superior to other arithmetic with respect to accuracy, costs, and speed.

The authoritative reference on the theory and design practice of computer arithmetic.

The role of arithmetic in datapath design in VLSI design has been increasing in importance over the last several years due to the demand for processors that are smaller, faster, and dissipate less power. Unfortunately, this means that many of these datapaths will be complex both algorithmically and circuit wise. As the complexity of the chips increases, less importance will be placed on understanding how a particular arithmetic datapath design is implemented and more importance will be given to when a product will be placed on the market. This is because many tools that are available today, are automated to help the digital system designer maximize their efficiently. Unfortunately, this may lead to problems when implementing particular datapaths. The design of high-performance architectures is becoming more compli cated because the level of integration that is capable for many of these chips is in the billions. Many engineers rely heavily on software tools to optimize their work, therefore, as designs are getting more complex less understanding is going into a particular implementation because it can be generated automati cally. Although software tools are a highly valuable asset to designer, the value of these tools does not diminish the importance of understanding datapath ele ments. Therefore, a digital system designer should be aware of how algorithms can be implemented for datapath elements. Unfortunately, due to the complex ity of some of these algorithms, it is sometimes difficult to understand how a particular algorithm is implemented without seeing the actual code.

The subject of this book is the analysis and design of digital devices that implement computer arithmetic. The book's presentation of high-level detail, descriptions, formalisms and design principles means that it can support many research activities in this field, with an emphasis on bridging the gap between algorithm optimization and hardware implementation. The author provides a unified view linking the domains of digital design and arithmetic algorithms, based on original formalisms and hardware description languages. A feature of the book is the large number of examples and the implementation details provided. While the author does not avoid high-level details, providing for example gate-level designs for all matrix/combinational arithmetic structures. The book is suitable for researchers and students engaged with hardware design in computer science and engineering. A feature of the book is the large number of examples and the implementation details provided. While the author does not avoid high-level details, providing for example gate-level designs for all matrix/combinational arithmetic structures. The book is suitable for researchers and students engaged with hardware design in computer science and engineering.

The original motivation for the development of digital computers was to make it possible to perform calculations that were too large to be attempted by a human being without serious likelihood of error. Once the users found that they could achieve their initial aims, they then wanted to go into greater detail, and to solve still bigger problems, so that the demand for extra computing power has continued unabated, and shows no sign of slackening. This book is an attempt to describe some of the more important techniques used today, or likely to be used in the near future, to perform arithmetic within the computing machine. There are, at present, few books in this field. Most books on computer design cover the more elementary methods, and some go into detail on one or two more ambitious units. Space does not allow more. In this text the aim has been to fill this gap in the literature. In selecting the topics to be covered, there have been two main aims: first, to deal with the basic procedures of arithmetic, and then to carry on to the design of more powerful units; second, to maintain a strictly practical approach. The number of mathematical formulae has been kept to a minimum, and the more complex ones have been eliminated, since they merely serve to obscure the essential principles.