Профессионально-коммуникативная подготовка студентов

The way information can be held in the computer’s RAM for display on the screen; this part of memory is sometimes called the display file.

The screen is divided into 24 lines of 48 characters, each character requiring eight bytes. The total RAM required for this screen is therefore 9 K, or 24·48·8 = 9216 bytes. The bits, in each group of bytes representing a character store the shape of the character itself, illustrated in Figure 1.

Fig. 1. The bits if eight bytes are used to represent a character on the screen, like this “E”.

BYTE 1 BYTE 2 BYTE 3 BYTE 4 BYTE 5

As an illustration of how the computer works, consider what happens when you type a letter (make it an “E”) on the keyboard. To begin with, the microprocessor will be running a program that is part of the ROM, checking the keyboard at regular intervals to see if a key has been pushed. At the same time, the VDU system is reading the screen RAM and displaying it on the screen as a picture. You press the key. The microprocessor checks which key has been pushed and, according to which ROM program it is running, may store the result in RAM somewhere. The microprocessor next uses a look-up table, stored in the ROM, to find out which codes are required to produce the letter “E”. It then checks other RAM locations to see whereabouts on the screen the letter should be printed, and finally writes the eight bytes that represent the letter into the relevant places in the screen RAM – note that the locations are not consecutive, but are actually 48 bytes apart. In practice the computer has to do a lot more than this, since the “E” may or may not have significance in the context of the program that is being run.

Above the screen RAM (usually), in the memory map, there is

20

another area of workspace that the microprocessor uses for storing temporary data like the printing position on the screen, the intermediate results of calculations, and all sorts of pieces of information that the system needs to remember. In our memory map, 2 K of RAM is allocated. This leaves just 29 K of “user RAM” available for programs, tiny by comparison with a personal computer, which would have anything upwards of 4 MB (megabytes) of user RAM. But it is enough to write a long and complex program for control purposes, which brings us to programming languages.

XIII. Work in pairs. Ask each other questions prepared at home.

XIV. Try to describe Figure 1. Don’t refer to the text.

XV. Work in microgroupes at the text “Computer programming languages”. Each microgroup prepares a report at one of the following topics: a) assembly language; b) low-level language; c) high-level language; d) compiled languages; e) interpreted languages; f) all types of languages.

COMPUTER PROGRAMMING LANGUAGES

The CPU of a computer – whether in a microcomputer or the largest mainframe – is programmed in binary code. It is almost impossible for humans to use binary code for programming. The nearest usable language to the binary code that the CPU needs is Assembly Language. Assembly Language instructions have a one-for-one correspondence to machine instructions: in other words, each Assembly Language instruction has an exact equivalent in binary code.

Assembly Language is not easy to learn, and it takes a long time to program a computer to do anything useful. An Assembly Language program to input two six-digit decimal numbers and divide one into the other, expressing the result as a decimal number, would take an experienced Assembly Language programmer a full week to write. Clearly there needs to be an easier way.

Assembly Language is known as a low-level language because it is close to machine language. Other computer languages are much

21

nearer to English, and are consequently easier to learn. Such languages can make it much simpler to program a computer, and are used wherever possible. Such computer languages are called high-level languages.

Programming languages are called “low level” when they are dose to machine language and don’t look like English. They are called “high level” when they are nearer to English.

There are two classes of high-level language: compiled languages and interpreted languages. Both translate something closer to English into a code understood by the CPU, but they do it in different ways. We will start by looking at the most widely used computer language of all, BASIC. The name is an acronym for Beginners Allpurpose Symbolic Instruction Code, and it was first used in the USA for teaching programming to university students, but has since been developed and extended until it can be used for a wide range of programming applications. BASIC is an interpreted language. A long and complex program (written in Assembly Language!) is kept in the ROM or RAM – this program is the BASIC Interpreter, and translates a program written in BASIC language into the binary code that CPU requires. One of the most popular compiled languages is still Pascal. The name is not an acronym this time, but is a tribute to Blaise Pascal, a seventeenth-century mathematician and philosopher. Pascal was designed at the outset to be a compiled language, and also to have a form such that its users are almost forced to write programs in an orderly, understandable way. Pascal compilers do not actually compile directly to machine code. Instead, they compile into an intermediate form called a P-code; the P-code is itself then run as an interpreted “language”, using a P-code interpreter! But the “interpreter” is generally called a translator in this context, and the result is something that runs a lot faster than an interpreted language, because all the hard part of the translation (Pascal to P-code) is done before running the program.

The speed of a compiled language is a function of the quality of the compiler – all else being equal, the better the compiler, the faster the object code will run. The skill in writing a compiler is in getting it to produce a relatively economic code. There are, of course, many different high-level programming languages. They are easier to write than Assembly Language, and they all run more slowly, for no compiler or interpreter has yet been written that can equal well-written As-

22

sembly Language for efficiency. Programming computers is something people can still do better than computers!

One of the oldest programming languages (and still going strong!) is FORTRAN (FORmula TRANslator). It is an excellent language for science and mathematics, and bears a close similarity to BASIC, which was developed from it.

Another language that is still widely used is COBOL (COmmon Business Oriented Language) which is good for producing lots of long reports, inventory and stock control, but too “wordy” for scientific work, graphic programs or mathematics. Pascal itself is a good gen- eral-purpose language, but is not particularly good for control applications. For heavy-weight applications – defence networks, for example

– languages like FORTH and Ada are used. For experiments in artificial intelligence (trying to make a computer behave like a person) a language called LISP is often used.

For applications programming where transportability (jargon for ease of translation for different makes of microprocessor and computer) is important, the programming language C, and its newer variants C+ and C++, are supreme. C++ is the language of choice for most commercial and scientific applications, because it is sufficiently lowlevel to provide a very good speed of execution, it puts detailed control of the machine into the programmer’s hands, and it is transportable.

XVI. One student from the microgroup makes a report.

A student from another group acts as his interpreter.

UNIT III

COMPACT-DISK SYSTEMS

I. Recognize the international words: compact-disk, system, synthesis, microprocessor control, spiral, stereo channel, decoder, synchronous, signal, laser diode, parallel, polarizer, concentric, project, component.

II. Memorise the following words.

23

III. Decode the following abbreviations: TTL, RAM, CMOS, DRAM, IGFET, ROM, EPROM, MOSFET, CPU, IC, VDU.

IV. Read and translate the text without a dictionary.

CD RECORDINGS

A CD player is a synthesis of a whole range of different aspects of electronics, without any one of which it couldn’t be made to work. CD players involve amplifiers, digital sound recording, microprocessor control, optoelectronics (including laser diodes) and servo control, just for a start.

CD players are, in theory and in practice, very complicated.

The compact disc itself contains a spiral recording (like the “old” vinyl records). The recording starts near the middle of the disc, and winds its way towards the outside edge of the disc over about 2000

24

revolutions in the playing area of 33 mm – a total maximum track length of over half a kilometre. Adjacent tracks are 1.6 ¸n apart, which means you could get about 30 of them across the width of a human hair. The spiral recording consists of a string of pits pressed into the aluminium foil on one side of the disc. The pits are about 0.5¸n wide and vary in length from about 0.8 to 3.6 ¸n, according to the sounds.

The pits represent binary data for the two stereo channels, sampled at about 40 kHz. A 16-bit system is used, in which each 16-bit binary number represents a 25 us sample of sound for each channel. Each sample has 65 536 possible values. For even the most sensitive ear, this system reproduces sound with perfect fidelity – not merely “hi-fi”, but perfect – along with zero background noise and no wow or flutter.

The disc whirls round at a speed that varies from about 200 to 500 revolutions per minute; the tiny embossed pits are “read” by a laser system that follows the track perfectly and senses the data at more than four million bits per second. It’s all very tiny and sounds as if it ought to be very near the limits of technology. And yet CD players are generally very reliable; it is unusual to experience a failure in the first two or three years of use under normal conditions.

V. Express agreement or disagreement with the following sentences using the phrases: you are not (quite) right, you are mistaken, you are wrong, I (can’t) agree with you. Consult the given above text. 1. The pits represent binary data for the two stereo channels. 2. The tiny embossed pits are read by the scanner. 3. CD players involve amplifiers, digital sound recording, microprocessor control, optoelectronics, and servo control. 4. CD players are very complicated in practice. 5. CD players are not reliable. 6. The compact disc contains a parallel recording. 7. The recording starts at the outside edge of the disc. 8. The disc whirls round a speed of about 500 revolutions per minute. 9. The spiral recording consists of a string of pits.

VI. Write down the corrected sentences from exercise V in successive order. Retell the text “CD recording” using these sentences as a plan.

25

VII. Read and translate the text using a dictionary. Ask questions on the topics: a) the reasons for coupling and verging bits; b) frames; c) CIRC; d) the rate of the system.

CD SIGNAL PROCESSING

Let us look at the audio data more closely. The system divides the data up into frames consisting of twelve 16-bit words; that is, twelve samples of the sound for each channel. However, the frame actually contains the following information.

1. 24 14-bit data symbols.

2.Eight 14-bit symbols used for error correction.

3.1 14-bit control signal.

4.A 24-bit sync signal.

5.102 coupling and merging bits.

Each 8-bit word (half a 16-bit sample for one channel) is coded into a 14-bit symbol on the disc. To this is added three extra “coupling bits” which are necessary to ensure that there are never two consecutive Is, which would confuse the recording. The 14-bit symbols are decoded into 8-bit words by the first decoder in the player, which does the error checking. To every 24 of the 8-bit words, the error-correction system adds the eight 8-bit words derived from the eight 14-bit error correction symbols, making 32 8-bit words in all.

A very complicated process called Cross-interleave ReedSolomon (CIRC) decoding now takes place. This greatly reduces the audibility of scratches or fingerprints on the surface of the disc by “spreading out” the error caused by any disc-reading fault over several samples, minimizing its effect. The CIRC decoder takes the 32 8-bit words and, during processing, subtracts four words of 8-bits, leaving 28 words of 8-bits. A second process removes another four 8-bit words, leaving 24 8-bit words, with any disc-reading errors spread inaudibly between them. These are assembled into 24 16-bit words, twelve for each channel. The sync signal consists of a 24-bit word and 3 coupling bits: it synchronizes the whole system to the rotation of the disc.

Each frame also contains a control signal – another 8-bit word coded as a 14-bit symbol. This holds the track and playing-time information which eventually finds its way to the display on the front of

26

the machine. The reasons for all those coupling and merging bits are twofold. Firstly, they are used to help equalize d.c. levels in the reading system (to make the electronics easier). Secondly, because the compact disc reads the transition between “pit” and “no pit” as a 1, and the gaps (either side or outside the pit) as a 0, strings of Is would make the pits too small to read.

When all the bits are added up they come to 588 bits per frame. With a sampling rate of 44.1 kHz (for each stereo channel) and a frame lasting 136 us there are about 7350 frames per second.

The system, therefore, has to read the disc and process information at the rather surprising rate of 7350·588 – 4 321 800 bits per second. Don’t worry, you don't have to remember all the above – but it shows just how complicated, even in theory, CD recording is.

VIII. Read and translate the text. Divide the text into logical parts. Entitle each part.

THE CD LASER SCANNER AND ITS CONTROL

The disc is read from underneath, through the transparent plastic. At the bottom of the pick-up is the laser diode, producing coherent light at a single frequency. The light is directed through a beam

splitter, then formed by a pair of collimating lenses into a parallel beam, passed through upolariser, and focused on the pits by an objective lens. The fact that the laser emits light at a single frequency enables simple lenses to be used.

Because the aluminium playing surface of the disc is shiny, the light is reflected back the way it came until it reaches the beam splitter. This directs the reflected light (but not the direct light from the laser) into a photodiode detector system. The detector responds to the pits with a signal that is decoded as Is or Os.

That is basically the way it works. Let’s refine it a little. The objective lens is mounted in a special “cell” which is a cross between a loudspeaker coil and a moving-coil meter movement. The lens can be moved up-and-down, or side-to-side. The whole assembly is light in weight, so it can be moved quickly if necessary. The up-and-down movement is used to keep the pits perfectly in focus on the detector. Without going into too much detail, the cylindrical lens introduces

27

astigmatism into the light reaching the detector. Astigmatism is an optical aberration which means that horizontal and vertical lines are focused in different planes – eyes sometimes suffer from it, and your optician corrects it with glasses that have a partially cylindrical lens to cancel out the effect.

When the laser is correctly focused, the photocells are all illuminated by a circular spot of laser light. Because of the astigmatism, an out-of-focus condition makes the circular spot become elliptical. The orientation of the major and minor axes of the ellipse depends upon which way the system is out of focus (too close or too distant). An error in one direction causes only cells 1 and 4 to be illuminated; the other way, and 2 and 3 are illuminated. These two different signals are amplified and used to move the objective lens up or down to maintain perfect focus. This part of the lens cell is like the middle of a loudspeaker, with a coil and magnet providing vertical movement in response to changing current through the coil.

The focus system is an example of a servo loop: any focusing errors cause the system to react to make a correction. In practice, small changes are being made all the time, as disc is unlikely to be absolutely flat. A second servo system controls tracking.

Compact discs are made to very close tolerances, and the tracks and the central hole would normally be concentric to within 50 ¸n. But with adjacent tracks only 1.6 ¸n apart, this is nowhere near good enough! There is a diffraction grating just above the laser. This is a flat piece of glass, engraved with very fine parallel lines. The grating breaks the laser beam into a number of divergent beams of different brightness; the (primary beam undiverged) is the brightest, but the secondary beams, deflected to the left and right, are also quite bright. The effect of the diffraction grating is therefore to project not just one but three spots of light on the disc’s surface.

In each diagram, the centre spot is the one produced by the primary beam, and the upper and lower spots are produced by the secondaries. The orientation of the diffraction grating relative to the disc is used to set up the correct positions.

How are the pits “read”? From where the laser is scanning them, they look like bumps, about 0.1 ¸n high, but they are actually quite flat, and have the same reflectivity as the rest of the surface of the disc. The detector notices them because the height of the bump (or the

28

depth of the pit if you are looking at it from the top) is such as to cause destructive interference of light at the laser frequency. Part of the scanning spot is “up” on the bump and part is “down” on the disc when a pit is passing the scanner. Light is reflected by each component in about equal proportions, but the length of the light paths differ by half a wavelength, causing destructive interference and a very considerable dimming of light reaching the photocells. The effect is not just simple “scattering” of light by the pits.

As the recording spirals out towards the edge of the disc, the pick-up has to be bodily moved to follow it. The tracking system recognizes how much correction it is applying to the objective, and when the correction reaches a threshold value a small motor starts and moves the pick-up out a little way. This is repeated as playback progresses. It is important to note that the motor does not have to position the scanner with absolute accuracy – the servos do the fine tuning. This means that the system can be built to “normal” precision engineering standards, and the price will be reasonable.

IX. Make up the abstract of this text.

X. Translate the given below texts in microgroupes. Exchange your translations and check them.

COMPLEX ELECTRONICS

The days of simple consumer electronic equipment are (in general) over. Most of today’s electronic gadgets, along with a vast array of machines that are now controlled electronically, range from complicated to extremely complicated. But they also tend to work reliably for long periods. Aerospace and defence equipment can be realty complex, and has the extra requirement of almost perfect reliability.

A lot of the continuing fascination of electronics is due to the fact that it is still evolving, with new ideas, new devices and new applications still pouring out of the laboratories and factories.

CONTROL OF THE DISC SPEED

There is actually a third servo system concerned with things me-

29