The Beginnings of the Microprocessor and Configurations of a Microprocessor System :
We will begin to learn the beginnings of the Microprocessor.
NOTE :
The microprocessor is a very complex electronic component produced on a silicon wafer of a few square millimeters.
On this chip are integrated logic functions, which required a 2 cubic meter computer in 1965. However, the microprocessor is not a microcomputer by itself.
The latter is a complete computer, of which the microprocessor is only the central part.
In the previous theories, we have already seen the structure of a microprocessor system.
1. - THE BEGINNINGS OF THE MICROPROCESSOR
The microprocessor was not born as a result of planned research but, it is the product of microelectronic technology and appeared in fortuitous circumstances.
It was in 1971 that the first large-scale integrated circuits (LSILarge Scale Integration) appeared, which contain up to 10 000 transistors. At the end of the same year, the first microprocessor also appeared : the INTEL 4004.
The electronic revolution of that time was linked to the development of NASA's space programs. Subsequently, government orders were reduced, which forced manufacturers to seek new outlets, particularly in computers and industrial automation.
In the early 1970, computers already present in the economies of different countries were built with SSI integrated circuits and with transistors. Certain programming languages had taken a certain extension : FORTRAN, COBOL, BASIC.
The notion of computer also penetrated the cultural sphere, which paved the way for the further development of microprocessors.
The 4004, prototype of the new generation of complex circuits, was satisfactory but was quite different from current microprocessors. It was especially suitable for mathematical calculations and was hardly suitable for general use.
The real general purpose microprocessor was born the following year ; it was INTEL's 8008 which was an 8 bits microprocessor.
The 8008 is a great success : for the first time, the designers had at their disposal a programmable component, which therefore offered great flexibility in its use.
Until that time, electronic circuits were made with the standard components available on the market.
A manufacturer could develop a complex integrated circuit only in a case of large production and at the request of a customer.
With the arrival of the microprocessor on the market, it became possible to use the same component to solve very different problems.
It is therefore this factor which then made it possible to market the microprocessor on a large scale.
Let's see, for example, how the microprocessor replaces traditional wired logic circuits.
Recall that, in a combinational circuit, the state of the outputs is determined only by that of the inputs. The input / output relationship constitutes the logic function of the circuit (Figure 1).
On the other hand, in a sequential circuit, the state of the outputs is a function both of the state of the inputs at the instant considered and of the state of the inputs at the previous instant.
Generally, a sequential circuit is controlled by a clock (Figure 2).
The microprocessor can replace any circuit, combinational or sequential (Figure 3).
The logic functions executed by a microprocessor system depend on the instructions stored in the memory. The longer the program, the more complex the logic functions. However, the execution of an instruction takes a certain time, so in some cases, the microprocessor will not be fast enough to perform the requested work.
This speed limit is the reason why microprocessors have not yet replaced all of the traditional circuits.
Traditional logic is still used for the execution of simple tasks (Figure 4).
Figure 4 above, on this card, more than 90% of the functions are performed by 5% of the components ; 95% of circuits perform only marginal tasks.
Figure 5 shows how programming a microprocessor can replace a combinational logic circuit. The function to be solved is as follows :.
You observe that the values present on the inputs A, B, C and D are treated at the same time in the case of the logic circuit, while they are treated one after the other in the case of the use of a microprocessor.
The 8008 microprocessor was so successful that the company INTEL undertook to develop a more efficient microprocessor. This is how the 8080 appeared, which became the most widespread microprocessor.
Meanwhile, other microprocessors appeared on the market : the 6800 models from Motorola, MM78 from Rockwell, 2650 from Signetics, F8 from Fairchild.
Microprocessor systems, compared to traditional circuits, have two advantages: :
1) - They use a limited number of integrated circuits, which allows :
a reduction in volume and a reduction in cards,
a decrease in the power consumed,
a reduction in connections therefore an increase in reliability,
a low price.
2) - They are programmable ; you just have to change the program to perform different tasks. In addition, in the event of errors in the project or simply, in the event of modifications, it is sufficient to make changes at the program level only.
In general, in the project of a digital system, one can choose from four methods :
a) - This constituted by the assembly of the various logic integrated circuits (traditional method). The system thus produced is often called wired logic or discrete logic.
b) - The one that uses a microcomputer (Personal Computer) where the logical function is carried out by a program written in high level language (advanced language).
c) - The "Custom Chip" method where the project is developed in close collaboration between the client and the manufacturer of integrated circuits.
d) - The one that uses a microprocessor ; in this case, as for the microcomputer, it is a specific program which performs the logic function. This program is written directly in machine language.
The table in Figure 6 makes it possible to compare these four methods mentioned.
The microprocessor is essentially constituted by the central processing unit (CPU) integrated in a single box (chip). This central unit can be more or less complex depending on the technology used (bipolar logic or MOS logic).
On the other hand, technological progress allows an increasingly high level of integration (Figure 7).
The first 4 bits microprocessors, derived from 4004, were generally used to replace logic circuits in industrial applications.
Those of the next generation, at 8 bits, were and are still used for arithmetic calculations and for logical processing.
The most recent microprocessors, 16 bits or 32 bits or even 64 bits, constitute a new stage in logic processing and in general uses, in programming (Figure 8).
The evolution of microprocessors is currently continuing. Ultimately, a slowdown is expected in the hardware design of the microprocessor.
The evolution will then be mainly at the software level.
There are already over a hundred types of microprocessors on the market. Most have common characteristics which are :
Von Neumann architecture (Programmability),
data organized in 8 or 16 bits words,
instructions consisting of 1, 2, 3 or 4 words,
8 or 16 bits external data bus,
16 or 24 bit address bus,
control signals entering or leaving the microprocessor.
Figure 9 shows the housing of a typical 40 pins microprocessor.
One of the limits to the technological development of microprocessors is the number of pins.
In recent years, 40 pins packages have dominated the market. More recently, boxes with 50, 64, 68 and even 80 pins have appeared. Figure 10 shows a type of 68 pins package.
The microprocessors (4, 8, 16, 32, 64 bits) are mounted with other integrated circuits on a card which constitutes the CPU card. There are four different types of system configuration depending on the microprocessor used.
a) - The Figure 11 represents the "standard" configuration, common to most of the microprocessor systems.
The microprocessor contains the traditional functions of the CPU. The single read (ROM) and read / write (RAM) memories, as well as the PIO, are external.
The PIO (Parallel Input / Output) allows communication between the computer and external circuits.
The microprocessor system circuits are connected by buses (external buses).
In Figure 11, each bus is represented by a single link, but in fact there are several. You notice that there are 3 buses.
- The data bus, bidirectional which routes data between the microprocessor and the external circuits (8 lines for an 8 bits microprocessor).
- The unidirectional address bus allows the microprocessor to address an external circuit.
- The control bus, consisting of the lines which carry the system synchronization signals.
In Figure 11, the clock circuit is external to the microprocessor. In newer microprocessors, however, it is internal ; only the quartz is outside. In addition, this type of microprocessor contains all of the system components.
b) - This leads to a second type of configuration (Figure 12) in which the microprocessor is a microcomputer by itself.
The microprocessor contains the CPU, buses, clock circuit, ROM, RAM, as well as other circuits (timer, analog / digital converter).
Figure 13 shows the integrated circuit of such a microprocessor.
You can observe the different zones (ROM, RAM, ALU ...).
The memory areas (ROM and RAM) are very regular while the areas of the input / output circuits are irregular.
c) - A third type of configuration is shown in Figure 14. It is a microcomputer halfway between the standard configuration and the "single chip" configuration.
The microprocessor used has more possibilities than a standard microprocessor. In particular, it can communicate directly with input / output devices using a bus.
The second integrated circuit contains ROM, RAM and input / output control circuits.
These two circuits thus constitute a microcomputer alone.
A microprocessor is chosen according to the application in which it is inserted.
Single chip microcomputers are used for simple industrial controls or for household appliances (ovens, washing machine, etc.). The mass production of these microcomputers (more than 100 000) reduces costs. The programs are thus stored directly in the ROM of the microprocessor.
2 circuits microcomputers are used for more complex applications, but which require limited memory.
Standard microprocessors are used for the most complex tasks.
d) - The last type of configuration is shown in Figure 15.
It is a microcomputer made up of integrated circuits called bit slice ("slice" = tranche).
Each "bit slice" (N° 1, N° 2, N° ...) processes a group of bits (generally 4 but also 8 or 16 bits).
The set of "bit slices" constitutes the processor. The latter can thus process words of 8, 16, 32 or 64 bits and even more.
A specialized circuit contains a micro-program which interprets the instructions and generates the control signals necessary for the execution of these instructions.
These "bit slice" processors have been designed to increase the speed of work. The low power Schottky bipolar technology is also used for this.
The downside is that the processor takes up more space than the three configurations seen before.
We will now study the internal architecture of the microprocessor (still in the same lesson, 4th theory).
Microprocessor system configurations
.
Click here for the next lesson or in the summary provided.
Top of page
Next Page 