C. Gordon Bell

16 – The Evolution of the PDP-11

Publisher Summary This chapter focuses on the evolution of the PDP-11. A computer is not solely determined by its architecture; it reflects the technological, economic, and organizational aspects of the environment, in which it was designed and built. The chapter discusses the nonarchitectural design factors: the availability and price of the basic electronic technology, the various government and industry rules and standards, the current and future market conditions, and the manufacturing process. The first weakness of minicomputers was their limited addressing capability. The biggest mistake that can be made in a computer design was that of not providing enough address bits for memory addressing and management. The PDP-11 followed this hallowed tradition of skimping on address bits; however, it was saved by the principle that a good design can evolve through at least one major change. A second weakness of minicomputers was their tendency to skimp on registers. This was corrected for the PDP-11 by providing eight 16-bit registers. Later, six 64-bit registers were added as the accumulators for floating-point arithmetic.

6 – The PDP-1 and Other 18-Bit Computers

Publisher Summary This chapter discusses the PDP-1 and other 18-bit computers. The Digital Equipment Corporation (DEC) System Modules that formed the basis of the PDP-1 were directly patterned after the circuits of the TX-0 and the TX-2 computers at Massachusetts Institute of Technology. Because the hardware for the PDP-1 was relatively inexpensive, DEC could afford to stock an ample supply of basic modules for building special interfaces. Constructing interfaces and specialized hardware was relatively easy compared to modern day hardware design. The Memory Test Computer (MTC) packaging, circuits, and toggle switches influenced the subsequent TX-0 design. The MTC packaging used various standard radio relay racks and had a somewhat homely appearance. The MTC circuits used significantly smaller modules than those in Whirlwind and used a gated pulse delay line clock for control rather than the synchronous clock used in Whirlwind. In addition, MTC used a dc bus for gating registers to one another that was carried out on an open-wired bus that ran the entire length of the computer.

1 – Seven Views of Computer Systems

Publisher Summary A computer is determined by many factors, including architecture, structural properties, the technological environment, and the human aspects of the environment, in which the computer was designed and built. Computer engineering is the complete set of activities, including the use of taxonomies, theories, models, and heuristics, associated with the design and construction of computers. This chapter presents seven views of computers and the process that molds their form and function. There are at least five levels of system description that can be used to describe a computer. Each level is characterized by a distinct language for representing the components associated with that level, their modes of combination, and their laws of behavior. The components of the circuit level are resistors, inductors, capacitors, voltage sources, and nonlinear devices. The behavior of the system is measured in terms of voltage, current, and magnetic flux. Above the circuit level is the switching circuit or logic level. The level above the switching circuit level is called the register transfer (RT) level. The last level is called the processor-memory-switch (PMS) level. This level, which gives only the most aggregate behavior of a computer system, consists of central processors, core memories, tapes, disks, input/output processors, communications lines, printers, tape controllers, buses, teleprinters, and scopes.

2 – Technology Progress in Logic and Memories

Publisher Summary This chapter explains the way in which technology progress, particularly in the areas of semiconductor logic, semi-conductor memories, and magnetic memory media, have influenced progress in the computer industry and have provided choice and challenge for computer design engineers. A single transistor circuit, performing a primitive logic function within an integrated circuit, is among the smallest and most complex of man-made objects. Alone, such a circuit is intrinsically trivial; however, the fabrication process required for a set of structures to form a complete integrated circuit is complex. Without well-defined functions such as addition and multiplication semiconductor suppliers cannot provide high density products in high volume because there are few large-scale, general purpose universal functions. The alternative for users is to interconnect simple logic circuits gates; however, that does not permit efficient use of the technology, and the cost per function remains high because the printed circuit board and integrated circuit packaging costs limit the cost reduction. The function performed by a chip is dependent on the number of gates that can be placed on a chip. Thus, density in gates per chip is the single most important parameter determining chip functionality.

7 – The PDP-8 and Other 12-Bit Computers

Publisher Summary This chapter discusses the PDP-8 and other 12-bit computers. Like PDP-5, the PDP-8 is a single-address 12-bit computer, designed for task environments with minimum arithmetic computing and small primary memory requirements. Typical of these environments are process control applications and laboratory applications, such as controlling pulse height analyzers and spectrum analyzers. In addition to the originally envisioned applications, the PDP-8 is used for innumerable other applications. One of the most interesting is message switching. The PDP-8 message switching hardware assembles characters by bit sampling, checking the status of teleprinter lines at five times the anticipated bit rate to accurately recover data. New technology can be utilized in the computer industry in three ways: (1) lower cost implementations at constant performance and functionality, (2) higher performance implementations at constant cost, and (3) implementation of new basic structures. Of these three ways, the PDP-8 Family has primarily used lower cost implementations of constant performance and functionality.

3 – Packaging and Manufacturing

Publisher Summary This chapter focuses on packaging and manufacturing of computers. Packaging is one of the most important and most complex elements of computer engineering. The importance of packaging spans the size and performance range of computers from the super computers to the pocket calculator. The importance of packaging is equaled only by its complexity. The complexity stems from the range of engineering disciplines involved. Packaging is the complete design activity of interconnecting a set of components via a mechanical structure to carry out a given function. To package a large structure such as a computer, the problem is further broken into a series of levels, each with components that carry out a given function. The product cost is determined by the price of the components and the manufacturing process; the life cycle cost includes the purchase price, the operational costs, and service costs. For production, machines must be easy to assemble and test, repair must be rapid, engineering changes must be introduced smoothly, and the production line cannot be held up because of shortages of components—all parts of traditional manufacturing considerations. Product announcement usually occurs during the design maturity testing period; however, it can occur any time—often as early as when the breadboard works or as late as the first customer shipment, depending on the marketing strategy.