The memories of a man in his old age are the deeds of a man in his prime.
In theory, there's no difference between theory and practice; in practice, there is.
It's all very well in practice, but it will never work in theory.
—French management saying
An ounce of action is worth a ton of theory.
A theory must be tempered with reality.
If the facts don't fit the theory, change the facts.
In this chapter we give a brief history of the application of asynchronous design. An in-depth study is given for one recent successful asynchronous design, Intel's RAPPID instruction-length decoder. Issues in performance analysis and testing of asynchronous designs are also discussed. The chapter concludes with a discussion of the synchronization problem. Although this problem may have prevented the commercial application of RAPPID, it may ultimately be the problem that makes asynchronous design a necessity.
Since the early days, asynchronous circuits have been used in many interesting applications. In the 1950s and 1960s, asynchronous design was used in many early mainframe computers, including the ILLIAC and ILLIAC II designed at the University of Illinois and the Atlas and MU-5 designed at the University of Manchester. The ILLIAC and ILLIAC II were designed using the speed-independent design techniques developed by Muller and his colleagues. The ILLIAC, ...