Widely used in both engineering and computer science departments (see [1]), this venerable text celebrates its 25th birthday this year with its sixth edition. By now, two generations of students have learned the principles of computer organization and design from it.
Used books from last year won’t do for this year’s classes, alas. Students will have to buy the new edition for more reasons than just keeping the problem numbers straight. Just about every perishable topic has been updated: newer circuits are included, the references have been refreshed (about three-quarters of the citations in the bibliography are new), and there are photographs of current hardware. (Don’t let the bouffant hairdos throw you though--some of the old pictures are still here.)
This edition has wider pages (and fewer of them), two-color printing for the first time, and a much better graphics layout. When the text refers to a table or diagram, the chances are you won’t have to turn too many pages to find it. Each chapter now begins with an explanatory paragraph or two and a statement of objectives, and each ends with a summary.
What has not changed is the clear, readable, and methodical exposition of the material. Chapter headings are as follows:
(1) Computer Operation,
(2) Number Systems,
(3) Boolean Algebra and Gate Networks,
(4) Logic Design,
(5) The Arithmetic-Logic Unit,
(6) The Memory Element,
(7) Input-Output Devices,
(8) Buses and Interfaces,
(9) The Control Unit, and
(10) Computer Organization.
Appendices, the bibliography, and answers to odd-numbered questions are at the back of the book. The book is as free from typos and other mistakes as six editors can make it. I noticed two errors which are trivial since they fall outside the principal scope of the book; nevertheless, they seem to be common misconceptions. For the record, the early machine EDSAC was developed at Cambridge, not at the University of Manchester, and the UNIVAC was not really “the first commercially available electronic digital computer.” LEO and HEC1 got there sooner. (For an interesting comparison, see <CR> 24, 7 (July 1983), Rev. 40,446.)
My other quibbles are the following: the paper and binding seem to be inferior to those of previous editions; I don’t like the OCR-style block letters which appear gratuitously in titles and elsewhere; and I found the brief section on programming too cursory to be of any value.
In any event, none of this detracts from the rest of the text. After the first assigned reading (Chapters 1 and 2), I recommend the book without reservation. Since 1960, this text has always provided the best available coverage of the material, and it gets better with each new edition.
