Institut National D'Electricite Et D'Electronique Boumerdes, Algeria
FIRST INELEC CONFERENCE
DIGITAL TRENDS IN THE 80'S
28 JUNE - 2 JULY, 1980
The Evolution of Digital EngineeringBy
Dr. George K. Kostopoulos
Institute National d'Electricite et d'Electronique
ABSTRACTEngineering is the vehicle by means of which Society is transported along the road of material progress. For such progress to be lasting, it is imperative that it be accompanied by similar progress in other, equally important aspects of life, namely, cultural and social. For the realization of this balanced progress, the engineer needs to receive himself a similarly balanced education. He should be taught that his role is not merely that of a master of numerical solutions, but that of a Social Component, who will identify and fulfill the needs of his environment and establish new goals for it. The engineer must first satisfy the fundamental needs of his Society before introducing glamorous foreign technologies. Such technologies introduced into the Society will be initially impressive giving a false feeling of success. Soon, however, will make the already existing technological and social gap even more obvious and raise questions as to the need for such advanced technologies. In a country's modernization process there will always be differences regarding the ordering of the priorities, often aggravated by the existing economic system. However, under all circumstances, in this decision process the engineer, and ideally the engineer-economist, should be the central figure identifying the cost effectiveness and the resulting social benefit of the various proposed projects. Consequently, the engineering education in the developing world should not be a carbon copy of that of the industrialized world, but a tailor made one meeting the unique requirements of the specific environment. As for the emphasis, it should be placed on the uniform, step by step, and omnidirectional development with realistic goals strictly based on the available material and human resources.To talk about the evolution of something we need to first identify its starting point. So, when did Digital Engineering start? Where is its origin in the time domain? Did it start with the fire signals of the African Indians? Or, did it start with the drum signals of the African Jungle? Or, did it start with the Ancient Philosophers of Logic? Is Digital as opposed to Analog a human invention? But, when did it start? It must have a beginning. Some origin. Well, the origin of Digital Engineering is the Quantum. It is the Atomic Theory itself. The orbits of electrons about their nucleus are discretely are discretely fixed. They are not linearly proportional to the energy of the electrons. They are the result of an analog digital conversion. In other words, the origin of Digital Engineering is placed on the time axis at t=0. At the beginning of Nature. And, if the Creator needed time to plan the Universe, then being part of his planning Digital Engineering started at t < 0. Therefore, the digital way - the discrete quantities - is inherent in every natural process. As a result, it was a matter of time when Digital Engineering would come out of the invisible atoms and become the dominant philosophy - the dominant way of life - in the World of Electrical Engineering. Digital Engineering will eventually serve all aspects of human life, the same way the Quantum Theory serves all aspects of Nature. During the centuries, the Logic Theory was developed and placed in the philosophical archives. The Binary Numerical System was defined and placed in the Mathematical archives. And the ON/OFF signals were looked upon as primitive, unworthy or engineering value. However, in less than half a century the Logic Theory, the Binary System and the ON/OFF Signals became the Holy Trinity of a new most promising engineering field. The rapidly expanding Digital Engineering. Since the beginning of the century, and even until my college years, electromechanical relays were the fundamental components of what was then known as Logic Design. As a matter of fact, my senior undergraduate project was a General Purpose Logic Equation Tester that used dozens of telephone type relays. As that time vacuum tube computers, very few in existence, were looked upon as state-of-the-art systems. Very valuable machines. I must say, as valuable as they are today to museums and antique collectors. As for the applications of Logic Design, they were limited, restricted by the available components, which were big, slow, heavy; with power requirements that were difficult to meet. In the 50's the availability of the transistor meant a break-through. Logic Design was elevated. It became Digital Engineering. A field with its own mathematics, theories, components, and engineers. Later in the 60's the development of the Integrated Circuits signaled the real beginning of Digital Engineering. And, that's where its modern origin lies. As for the 70's, "digital" has become the magic word. In the 80's, there will be a new magic word abbreviated M2PS, standing for multi- microprocessor system. We have now reached the point where the inter-micro-computer communication and allocation of tasks - the taxonomy, as it is known has become our major concern. The abundance of digital integrated circuits, in types and quantities, standardized and manufactured all over the world, has made previously unthinkable applications realizable. The digital watch and pocket calculator will undoubtedly be our pride, if we ever visit an inhabited planet. How has this progress reached that level? Let's walk through it. At the component level, the electromechanical relay and the vacuum tube were replaced by the transistor. The transistor was replaced by the custom integrated circuit. The custom IC by the Standard RTL, the RTL by the DTL, and the DTL by the TTL and the Metal Oxide Semiconductor Logic. Along came the Emitter Coupled Logic (ECL), which today is impractical in most applications because of its low noise immunity. But, it still hangs on for special research applications. Our great savior and best friend is now the Integrated-Injection Logic (I2L) with almost TTL speed and MOS density. From here on, it appears that higher density, lower power and higher speed will be steadily achieved, with TTL leading in speed, OMOS in density, and I2L competing with both. At the subsystem level, memories, the most essential part of any substantial digital system, have also evolved in density and performance with numerous types being available; and they will remain numerous because each memory type serves a slightly different price/speed ratio. It started as a single flip-flop chip, storing one bit, progressed into a storage register, and now, we are talking about a Random Access Memory (RAM) chip with 16K bits. Not far back, say fifteen years ago, a four bit register was a state-of-the-art device. Undoubtedly, the Charge-Coupled Devices (CCD) advanced Digital Engineering by providing a link between the random access memories and the magnetic memories. Seven years ago, I was the proud user of a 512-bit dynamic shift register. Today, 64K-bit devices are not uncommon. In the area of bubble memories, engineers have placed very high hopes on the Josephson Memory Cell, which has achieved the ultimate of efficiency in magnetic storage. Storage of one bit of information requires only one quantum of magnetic flux. In a parallel path we have the development of the microprocessors the arrival of which has indeed, revolutionized digital design. Now, there are two distinct ways of digital system design. One is using conventional digital components, that is, gates, registers, counters, etc.; and the other is using one or more microcomputers. It started as an Arithmetic Logic Unit (ALU), it progressed into a Central Processing Unit (CPU), then into a Microprocessor (MP), and today, we have a single-chip Microcomputer with its CPU, Program Memory, Data Memory, multiple I/O interfaces and timing circuits; all on a single chip. In brief, the semiconductor chemists and physicists are providing Digital Engineering with practically unlimited infrastructure support, which will continue. They will be giving us devices that satisfy our imagination and challenge us to new more sophisticated applications. In the applications area the evolution of Digital Engineering is most evident. Consider the pocket calculator, for example. It started rather bulky, dim and with four operations, and in ten years it became slim, bright and scientific; even with an alarm clock that wakes you up to your own favorite tune. The last castle of analog engineering, which was communication, fell, and today, telecommunication and digital communication are practically one and the same. In the area of telephony the trend is irreversible. The process is slow, because of the tremendous volume of equipment that need to be replaced, but it has started. Progressively, all electromechanical equipment that comprise a telephone system will be replaced by digital ones. The reason being that digital are more cost effective. First to be replaced are the long distance voice transmission links, next the telephone switching apparatus, and last our most dear telephone set itself, which will accept and send voice signals in pulse code modulated (PCM) form. One can speak endlessly on how Digital Engineering is making life easier in practically all areas. We have spoken about the past and the present of Digital Engineering. The obvious question is: How about the future? Well, it appears that there are three forces that will play a major role in the further evolution of Digital Engineering. These are · the Very Large Scale Integration (VLSI) · the Josephson Junction, and · the Optoelectronics In the area of VLSI we will see chips with one million devices, which is translated into a 1M-bit RAM, or a single-chip 32-bit microcomputer, or a milti-microprocessor chip. We will also see a milestone reduction of the digital circuit supply voltage from 5 volts to either 3 or 1.5 volts, most probably to 1.5 volts. This will bring power dissipation to about one tenth its present level, allowing even higher IC densities. The new generation of microprocessors, which, in effect, will be the third generation, will be the same in bit size, but more powerful in architecture to allow a more extensive use of standard software packages. For all practical purposes we will have a minicomputer in a few chips. The second major force that will mold the future of Digital Engineering is the Josephson Junction Technology. This is a concept where the digital computer system operates at 4 degrees K, four degrees above absolute zero, submerged in liquid helium. The technology is based on magnetic flux control and requires very little of power, like half a microwatt per device. Its switching speed is, an otherwise unapproachable, 15 pico-seconds, able to perform 70 million instructions per second. The physical size can be as small as a 15 cm. cube for the computer system, with a refrigeration unit comparable to that of a home refrigerator in size. And, the third technological advance that will open up new roads in many areas of Digital Engineering are the optoelectronics. The optical disc is such an example. This is a 30 cm. disc, where by means of a lazer, data can be recorded and played back, and one such disc can replace up to 25 magnetic tapes. In the next few years we will see the video cassette market challenged by the video disc, that can store digitized video at a rate of 10 M bits per second. The video disc will capture a large portion of the market offering an alternative to video storage. Optical data transmission is another example, where date modulated light travels through plastic clear fibers for miles with practically no loss. When we think of the applications, it is hard to say which field is the greatest beneficiary of Digital Engineering. Maybe Computer Science, maybe Medicine, maybe Communications, maybe the consumer if we consider quantity. We are in a trend that continues. It will not stop. The input to the System is digital circuits - smaller in size and cost, but bigger in function - and the output of this System is more service to the user. Higher efficiency. More for less. Well, is this progress desirable? Is it all necessary? Is it good? The answer to these questions is NO. This progress is not desirable, neither necessary, not good ……. Unless, similarly advancing are the remaining aspects of the life of the individual, which this specific progress serves. The Society should not be enchanted by the glamour of the Advanced Technologies and neglect its real needs. Our contribution to a given social environment should not be providing luxuries to a few, but fundamental services to all. At this point, I would like to assure you of a bright future for Digital Engineering, but leave with you my hope that the brilliant example of the Evolution of Digital Engineering be matched by similar evolutions in other more fundamental aspects of human life.