Digital Arithmetic

Rating:
Summary: A good book by a good Professor
Review: I had taken the course "Digital Arithmetic" under Prof. Ercegovac in UCLA and had used the draft version of this book. Professor Ercegovac is extremely knowledgable and well-known in his area and I was amazed at how quickly he was able to answer all our doubts.

Now about the book. I have not read any other book on this subject except this one. So, keep that in mind while considering my review. This book is about efficient algorithms for arithmetic operations like addition, multiplication, etc and the ciruits for realizing these algorithms. The book deals with circuits at a high black-box level, so you don't have to be an expert in low-level circuits (in fact, it hardly needs any electrical enginering knowledge). I found the book to be pretty good. First the organization. The book starts with a brief review of various number systems, etc. I was pleasantly surpised to know that apart from the decimal, binary, n-ary number systems, we have what are called redundant number systems, etc. It was also fascinating to know how different number systems had different tradeoffs which made one number system better for certain operations (for eg. redundant number system is good for multiplication) but maybe not for other operations. Then the book discusses two-operand addition techniques. By the first few pages of the this chapter, you would have covered simple schemes like ripple carry adder, carry-save adder, etc which are normally covered in undergraduate computer architecture courses. Then the chapter discusses various other schemes for two operator addition. The next chapter deals with multioperand addition, which as the name implies, discusses different techniques for adding more than two operands. It is very amazing to see the authors use very clever tricks to reduce the complexity of the circuits needed to realize the operations. The later chapters discuss the more complicated operations of multiplication, division, square roots, etc. There is also a chapter on serial arithmetic. These later chapter may be a bit difficult to understand if you are planning to study them on your own and if you don't have much background in this area.

Overall, I found the book to be excellent. It has plenty of figures to explain the concepts and a number of exercises to test your understanding. The only minus point I found was that some sentences in the text were confusingly framed. This might be expected because the authors are not native English speakers. But, then I had used the draft version and I have been told that the print version is much better having gone through the editing phase. To conclude, after taking the course (with this textbook), I came out totally amazed by the fact that there is so much more to know than we can imagine about the "everyday operations" of addition, multiplication, etc.

Rating:
Summary: Every Bookshelf Should Have this Book
Review: One of the best books on arithmetic out there for professionals, researchers, students, and those generally interested in computer arithmetic. It explains everything in a simple and understandable manner.

I teach a computer arithmetic class and this is one of the best books out there. Its superb and super-highly recommended. The only thing to watch out for is that it doesn't cover topics in residue, logarithmic, and cryptographic arithmetic. However, I am sure the authors will include this in future editions.

Its an outstanding buy and something every VLSI and computer architecture should not be without.

Rating:
Summary: the comprehensive arithmetic text
Review: This is the most comprehensive, rigorous, yet readable treatment of computer arithmetic I have yet encountered. The authors are two of the leading authorities in the field. It is a must-own for anyone serious about the subject.