4. 1 Data Representation 1.2.1 BinaryNumbers Computers work in number base 2 which uses 2 symbols, 0 and 1 to represent a value. In computing systems, large numbers are expressed in terms of powers of 2 and use the following abbreviations: 2 1 has a decimal equivalent of 2 2 2 has a decimal equivalent of 4 2 3 has a decimal equivalent of 8 2 4 has a decimal equivalent of 16 2 5 has a decimal equivalent of 32 2 6 has a decimal equivalent of 64 2 7 has a decimal equivalent of 128 2 8 has a decimal equivalent of 256 2 9 has a decimal equivalent of 512 2 10 has a decimal equivalent of 1024 and is abbreviated to 1 kilo 2 20 has a decimal equivalent of 1,048,576 and is abbreviated to 1 Mega 2 30 has a decimal equivalent of 1,073,741,824 and is abbreviated to 1 Giga 2 40 has a decimal equivalent of 1,099,511,627,776 and is abbreviated to 1 Tera

7. 1 Data Representation 1.2.4 Hexadecimal Long binary numbers can be difficult to read correctly. Computers have memory addresses of 2 or 4 bytes long which give addresses of 16 or 32 bits. Hexadecimal is base 16 and organises the bits into groups of four. The conversion between base 2 and base 16 is very simple. Hex needs the digits 0-9 and letters A-F. E.g. 11010100010110010011001010010110 becomes 1101 0100 0101 1001 0011 0010 1001 0110 which in Hex is D459 3256

13. Data Representation 1.4 Graphics Most displays use Raster graphics – same as TV. Displays store images as a matrix of pixels in the refresh buffer. Separate images now stored in VRAM (Video RAM). VRAM represents the entire screen area and the term bit map is used to describe the one-to-one mapping of pixels in VRAM to pixels on the screen.

17. Data Representation 1.4.4 Greyscale A rudimentary greyscale effect provides a ’black’, ’white’ and two levels of ’grey’. As this comprises four different values we need two bits to represent each pixel (00 for black, 01 for darker grey, 10 for lighter grey and 11 for white ). As each pixel now requires twice as many bits, we will require twice as much memory for a given screen size as a black and white image. We can provide more levels of grey by allocating more bits to each pixel. By the time we have eight bits (one byte) to one pixel we can represent 256 different intensities. Monochrome displays are often clearer, especially for text than colour display. The requirement to use colour for such items as colour pictures and user interface issues, dictates that colour displays are more likely to be purchased.

19. Data Representation 1.4.5 Colour One colour can be represented by one byte giving 256 colours (GIF format). Monitors etc. have 3 primary (additive) colours, Red, Blue and Green. Other colours obtained from adding light. We use 8 bits for Red, 8 for Blue and 8 for Green which give us 256 x 256 x256 colours – over 16 million. We need 3 bytes to describe RGB coded colours. Codes can be used by a programmer to describe colours in Hex code.

21. 2 Computer Structure 2.1 An Introduction This unit on Computer Structure describes in detail the function of the component parts of a processor in the manipulation of data. This is extended to the methods of transferring data within a processor and between a processor and memory. The concept of a stored program is considered and the steps in the fetch-execute cycle to access and run programs. Memory types are considered, from registers to backing storage and how memory is defined and addressed.

26. 2 Computer Structure 2.2.2.1 The structure of the CPU (a) Memory Processor Control unit ALU Registers, A, MAR, MDR, PC, SP Address bus – 1 way Data bus – 2 way Control Bus Internal buses

28. 2 Computer Structure 2.2.3 The stored program concept All computers based on same basic design, known as the Von Neumann Architecture . Computers carry out tasks by executing machine instructions. A series of these instructions is called a machine code program held in main memory as a stored program , a concept first proposed by John Von Neumann in 1945. Central Processing Unit (CPU) fetches, decodes and executes the machine instructions. By altering the stored program it is possible to have the computer carry out a different task.

29. 2 Computer Structure 2.2.4 The fetch execute cycle To execute a machine code program it must first be loaded, together with any data that it needs, into main memory (RAM). Once loaded, it is accessible to the CPU which fetches one instruction at a time, decodes and executes it at electronic speed. Fetch, decode and execute are repeated until a program instruction to HALT is encountered. This is known as the fetch-execute cycle .

32. 2 Computer Structure 2.2.8 Computer Components and Their Function The components of the CPU and the connections to devices that are external to it are shown.

38. 2 Computer Structure 2.4 Central Processing Unit Central Processing Unit. The CPU coordinates and controls the activities of all other units in the computer system. It executes program instructions and manipulates data in accordance with the instructions. It uses a standard architecture composed of the following three components: Arithmetic and logic unit (ALU); Control unit; Registers. All three components work together to form the processor.

39. 2 Computer Structure 2.4.1 Architecture of the microprocessor We will now study the internal architecture of the microprocessor (CPU) itself. Because of the stored program concept , we must consider the relationship between the CPU and memory. This is a diagram of a fairly typical microprocessor design, showing the internal structure of the CPU and its relationship to the memory of the computer.

41. 2 Computer Structure 2.4.2 Accessing Memory (2) To read data from memory , CPU places the address of the memory location into the MAR and activates the memory-read control line of the system bus. This will cause the required data to be transmitted from memory via the data bus to the MDR; To write from the CPU to memory , the CPU places the data to be written in theMDR; the address of the memory location where they are to be written is placed in the MAR; and the memory-write control line is activated. The MAR and MDR registers have a large part to play in the fetch-execute cycle.

68. 4.2 – Input & Output Devices 4.2.7 Multiscan Monitor The CRT is the basis of most visual display technology. The screen is arranged as a series of lines of dots and each dot is made up of three small areas of red, green and blue called a triad. The intensity of light shone on each triad determines the actual colour of the pixel. The picture is redrawn between 50 and 100 times a second. This is the refresh rate. A monitor which operate at different refresh rates is known as a multiscan or multisync monitor. The refresh rate is controlled by the video adapter. Screen resolution is quantified by the dot pitch, the distance between the dots on the screen. Typically between 0.28 and 0.38mm, corresponding to 100 to 70 dpi.

71. 4.4 Buffers and Spoolers 4.4.2 Spooling When large amounts of data are to be sent to a peripheral device, or when the peripheral is shared across a network then spooling is a preferred method of compensating for the difference in speeds of the processor and the peripheral. Spooling involves the input or output of data to a tape or a disk. This, for example, allows output to be queued from many different programs and sent to a printer by a print spooler (special operating system software). The print spooler stores the data in files and sends it to the printer when it is ready, using a print queue . Once the data has been printed it is deleted from the storage device.

72. 4.5 – Storage Devices 4.5.1 Magnetic Magnetic storage devices include hard disks, floppy disks, Zip disks and magnetic tape. They are called magnetic storage devices because their recording surfaces are coated with a material that responds to magnetic fields to enable data to be stored. Storage devices can be fixed or removable. Removable storage devices allow the user to disconnect the device and physically transport data from one computer to another. Varieties of removable devices include the Iomega and Syquest hard disks and Jaz cartridges.

75. 4.5 – Storage Devices 4.5.1.2 Magnetic Tape Storing data on tapes used to be the only solution to backing up hard disks of large capacity. Now, with large, removable magnetic disks and optical CR-RW technology, this is no longer the case. However, removable storage media is comparatively expensive, costs 10 times tape. Tape, therefore, still has the edge in this market. Tape is read and written on a tape drive. Data is written to tape in blocks with inter-block gaps between them. A single operation writes each block Data is stored on magnetic tape as magnetised regions on the surface of the tape induced by the magnetic recording head. To read data, the tape passes under the read/write head and the stored magnetised regions produce very small voltages in the head, leading to a current in the head coil. This current can be analysed to give a representation of the stored binary data.

76. 4.5 – Storage Devices 4.5.1.2 Magnetic Tape Capacity Magnetic tapes have large capacities, reaching up to several gigabytes and come in a variety of sizes and formats. Since their introduction, tape drives have passed through many stages of improvement with extremely reliable Digital Audio Tape (44.1 kHz, 16-bit record and playback DAT) drives representing the current state of the art. A 4mm DAT tape can now store up to 24 Gbytes of data! Access Tapes are sequential access devices. Accessing data on tapes is therefore much slower than accessing data on disks. They are not suitable as storage media for applications where data needs be used regularly - where a disk is a more appropriate medium. Because tapes are so slow, they are generally used only for long-term storage and backup.

110. 5 Networking 5.4.1 Network Topology - Bus 4 Factors Affecting Performance 4.6. Network Topology - Bus Bus Topology- easy to expand and cheap to set up. e.g. Ethernet in school or college. Data Security – data encryption methods used. High collision rates requiring re-transmission Bandwidth – available bandwidth shared amongst all stations accessing network. Data compression used. Reliability – fault in one station has no effect on rest. Cable fault will lose all that section. Cost – relatively cheap

111. 5 Networking 5.4.2 Network Topology - Star Data Security – higher security as data routed only to the computer that is to receive it. No collisions. Reliability – if a link fails then only that station is off the network. Failure to central controller is fatal. Cost – can be quite expensive due to high cost of cabling. Popular in small self-contained networks as not too expensive (small office). All nodes connected to one central node that routes traffic to the appropriate place.

112. 5 Networking 5.4.3 Network Topology - Ring Ring Topology Similar to Bus in many respects with similar security problems. Control system in charge of transmissions and stations guaranteed access to transmissions. Collisions avoided by use of a token Additional expense for control s/w and system. May have to wait turn to transmit. Network down to add station, but few if any crashes.

113. 5 Networking 5.4.4 Network Topology - Mesh Mesh Topology Fault in one cable does not affect network. Multiple transmissions Lots of wiring Expensive Excellent performance

124. 6 Using Networks 6.4 Technical Factors Affecting Communications The technical factors which have led to the growth of computer networks have emerged in parallel with t he economic factors which have driven the research into networking technology. As the economic demand or networking technology has grown, the trend has been for equipment prices to fall and performance to increase. Although still in its infancy, the development of wireless networking is likely to follow the same pattern.

142. 6 Using Networks 6.5.3 Misuse of Networks – Network Etiquette