(1971) for structure and behavior, knowledge of these, languages (notations) is not required. We have avoided using a register transfer language for defining behavior. We use instead the flowchart, with ISP language expressions. This method is equally formal and precise, yet shows 'the structure and parallelism without burying the reader in the language control syntax. We do invite the reader to describe and design RTM systems in his favorite (1 dimensional) register transfer language, as an exercise. Alternatively, he should simulate his design in a conventional language (e.g., Fortran). In a few instances, we have described systems in pure ISP language.
The book is aimed primarily at people who wish to design specific small digital systems -- either to learn about digital technology or to accomplish a particular task. The book is specifically designed to be used as either a primary or a secondary text in undergraduate and graduate courses on logical design. It can also be used for self study by students in departments of electrical engineering. and computer science. It is especially suited for the latter, if the department does not provide any courses that descend below the programming level to tell how digital systems really operate. Although not a requirement, a knowledge of a programming language will be helpful in reading the book.
When the book is used as a primary text, at the sophomore or junior level, as a first course, some minor amount of supplementary material on number representation and Boolean algebra may be useful. After the first four chapters, the book becomes topical and can be read in any order. We strongly recommend the style of design and analysis given in Chapter 4, and several weeks can be spent studying this chapter.
The book is also designed specifically for self study by engineers and scientists of all persuasions (civil, mechanical, chemical,..., physics, chemistry, biology,..) who deal with laboratory and industrial processes that need monitoring and control. As this book demonstrates, the design of RT-level systems is no more difficult than programming, the construction of working systems is no more onerous than debugging programs,, and the cost of such systems is low enough to be cost-effective for use with all manner of laboratory instrumentation, industrial instrumentation, and production processes. The performance that can be delivered in the way of useful bits processed per second per dollar is, for the right task in the right place, phenomenal. The design time appears to be cut at least a factor or 10, making it possible to design, wire, and test a simple computer or a desk calculator in about 20 hours.
If we appear to sell a little bit in the previous paragraph, that sour intent. There is a certain amount of missionary work to be done. The making of myths is a constant avocation of man. He establishes them continually as he packages his experiences in verbal formulas to make his world seem comfortable and familiar. They linger almost forever, being cut free from the old experiences that generated them. Some myths that cling to the computer field are: that hardware is much harder to construct than software, so always program rather than construct equipment; "that all digital systems are digital computers; that interfacing digital systems to the external world is complex and costly; that the only fit interfaces for digital systems. are punched cards in and printers out (recently modified to include Teletypes in and out); that digital systems are expensive facilities (i.e., capital equipment), rather than instruments or supplies,
vii