Fundamentals of Computer Organization and Architecture

Question 1

Fetch Execute Cycle

Answer 1

A continuous process carried out by the processor when running programs

Question 2

Two Types of Main Memory

Answer 2

RAM - Random Access Memory
- Temporary Storage and is volatile

ROM - Read Only Memory
- Generally only used for storing a limited number of instructions eg BIOS settings. Non-volatile memory.

Question 3

Buses

Answer 3

Data Bus - transfers data from the processor to and from memory, and to and from the I/O Controllers. The data bus is bi-directional/two way

Address Bus - Is used to specify a physical address in memory so that the data bus can access it. The address bus only goes in one direction - from the processor into the memory

Control Bus - Controls the flow of data between the processor and other parts of the computer. The control bus is bi-directional and sends control signals to registers, the data and address buses.

Question 4

Von Neumann Architectures

Answer 4

How Von Neumann architecture simply works is the instructions and data that comprise a program are both stored in the main memory and must pass through the same bus (data bus) in and out of memory.

Most modern PC’s use this technique.

https://bit.ly/2t6C1Up

Question 5

Harvard Architecture

Answer 5

The main difference is that separate buses are used for data instructions, both which address different parts of the memory.
https://bit.ly/2HQB5c2

The advantage of the Harvard architecture is that the instructions and data are handled more quickly as they do not share the same bus.

Harvard architecture is widely adopted on embedded computer systems such as mobile phones, burglar alarms etc where there is a specific use rather than being used within general purpose PCs.

Many such devices use a technique called Digital Signal Processing (DSP). The Idea of DSP is to take continuous real world data eg audio or video data, and then compress it to enable faster processing. Chips optimised for DSP generally have lower power requirements, making them ideal for applications eg mobile phones where power consumption is critical.

Question 6

Control Unit

Answer 6

Part of the processor that manages the execution of instructions.

Question 7

Register

Answer 7

A small section of temporary storage that is part of the processor. It stores data or control instructions during the fetch-decode-execute cycle

Status Register - Keeps track of the status of various parts of the computer - for example, if an overflow error has occurred during an arithmetic operation.

Interrupt register - Is a type of status register. It stored details of any signals that have been received by the processor from other components attached to it, for example, the I/O controller for the printer

Question 8

There are four registers that are used by the processor as part of the fetch-execute cycle

Answer 8

The current Instruction Register (CIR) stores the instruction that is currently being executed by the processor

The program counter (PC) stored the memory location of the next instruction that will be needed by the processor

The Memory Buffer Register (MBR) also known as the Memory Data Register (MDR) holds the data that has just been read from or is about to be written to main memory

The memory address register (MAR) stored the memory location where data in the MBR is about to be written to or read from.

Question 9

Fetch Decode Execute Cycle

Answer 9

Fetch
The address from PC is copied into MAR
Increment PC by one.
The address is then sent along the address bus to the main memory.
The contents of the address can now be transferred via the data bus into the MBR.
Transfer contents of MBR into CIR

Decode
The instruction in the CIR is now decoded by the decode unit

Execute

Question 10

Factors affecting processor performance

Answer 10

Clock Speed - Indicates how fast each instruction will be executed.

Bus Width - The number of bits that can be sent down a bus in one go.

Word Length - Computer systems may have a word length of 32 or 64 bits indicating that 64 bits of data can be handled in one pulse of the clock. Word length and bus width are closely related in that a system with a 64-bit word length will need 64 bit buses

Multiple Cores

Cache memory - Caching is a technique where instructions and data that are needed frequently, are placed into a temporary area of memory that is separate from main memory. The advantage of this is that the cache can be accessed much more quickly than main memory, so programs run faster. The key to this is ensuring that the most commonly used functions or data used in a program are placed into the cache.

Question 11

Interrupts

Answer 11

A signal sent by a device or program to the processor requesting its attention.

Since the fetch-decode-execute cycle is always in use, if an error was to occur or a device wants the computer to do something else then we need some way to grab the processor’s attention. The way to do this is to send an interrupt.

An interrupt is an additional step to the fetch-execute cycle. Now fetch, decode, execute and check for interrupt.

Interrupt service routine (ISR) - Calls the routine required to handle an interrupt.

To correct the interrupt, the processor places what it was currently doing onto a system stack, and using the ISR.

Priorities - A method for assigning importance to interrupts in order to process them in the right order.

Sometimes, the program that has interrupted the running of the processor is itself stopped by another interrupt. In this case, the processor will either place details of its current task onto a stack or it will assess the priority of the interrupts and decide which one needs to be serviced first. Assigning different interrupts to different priority levels means that the really important signals, such as a signal indicating that the power supply is able to be lost, get dealt with first.

Vectored Interrupt Mechanism
Details about how to deal with each interrupt are stored in a section of RAM. The start
address of each of these sequences of instructions is stored in a list so that, when an
interrupt occurs, the processor can use this list to find the code needed to deal with the
interrupt.

Question 12

CMP r1, #10

Answer 12

10 - Addressing Mode - the way in which the operand is interpreted

CMP - Operation Code/opcode

r1 - Operands - A value or memory address that forms part of an assembly language instruction

CMP r1, #10 means compare the value in register 1 with the value of 10

CMP r1, #10 ‘comment’

Question 13

Immediate and Direct addressing

Answer 13

Direct address - Using a direct address mode tells the CPU which address contains the data you want to access. So LDR r1, 100 would copy the data held in memory location 100 into register 1

Immediate address - Rather than loading the contents of a memory address, the immediate address method loads tha data directly. Therefore, the operand would have to be the actual number that you wish to use. A comand such as MOV r1, #10 would move the value 10 into register 1

Question 14

Types of Operation Code

Answer 14

PG 284 probably already mention in exam though.

Question 15

Charge Coupled Device (CCD)

Answer 15

In digital cameras, it is a sensor that records the amount of light received and convert it into a digital value.

Complementary Metal Oxide Semiconductor (CMOS) is an alternative technology that performs the same functions as a CCD.

Question 16

RGB Filter

Answer 16

Red, green, and blue filters that light passes through in order to create all other colours.

Question 17

How a camera takes a photo

Answer 17

When a photograph is taken the shutter opens and lets light in through the lens
The light is focused onto a sensor, which is usually either a CCD or CMOS.
The sensors are made up of millions of transistors, each of which stores the data for one or mor epixels.
As the light hits the sensor, it is converted into electrons and the amount of charge is recorded for each pixel in digital form.
With light, all colours can be created from red, green, blue (RGB). Therefore, to record colour, the camera will either have three different sensors, or use three different filters, one for red, one for green, one for blue.
The data is typicaly stored on removal storage devices, usually referred to as flash memory, which uses programmable ROM
Data is usually stored in compressed files, for example, TIFF, JPG or PNG
RAW files can also be generated, which are uncompressed and therefore contain all of the data from the original photograph.
This digital data can now be decoded and manipulated using specialised software.

Question 18

Barcode reader

Answer 18

A device that uses lasers on LEDs to read the black and white lines of a barcode.
How it works:
A light, usually an LED or laser is passed over an image.

Some form of light sensor is used to measure the intensity of light being reflected back. This is converted into a current effectively generating a waveform. This could be achieved using a photodiode or a CCD sensor in the same way as a digital camera.

White areas reflect most light and black areas the least, making it possible to use the waveform to distinguish the patterns of black and white bars.

The waveform is analogue and therefore needs to be converted into digital form using an analogue to digital converter.

The encoding will convert the black and white into binary codes eg black = 0 white = 1.

The signal is decoded into a form that can then be interpreted by software.

Question 19

Radio Frequency Identification (RFID)

Answer 19

A microscopic device that stored data and transmits it using radio waves, usually used in tags to track items.
Typical uses for RFID include tagging pets or livestock, or tracking products through a production line.

RFID works in the following ways:
The tag, which can be microscopically small, contains a chip, which contains data about the item and modem to modulate and demodulate the radio signals.
The tag also contains an antenna to send and receive signals.
Tags can be either active, which means they have their own power source in the form of a small battery, or passive, which means they will pick up electromagnetic power when they are in range of a RFID reader.
Signals and therefore data can be transmitted in both directions, using radio frequencies. This may be over a short or long distance depending on what the tags are being used for and how they are powered. The typical range of RFID tag is between 1 and 100 metres.
Tags may be used simply to track the physical location of the tagged item or the time may transmit data back.

Question 20

Laser Printer

Answer 20

How a laser printer works:
A rotating drum inside the printer is coated in a chemical which holds an electric charge.
The laser beam is reflected onto the drum and where the light hits the drum the charge is discharged, effectively creating an image on the drum.
As the drum rotates it picks up toner which is attracted to the charged part of the drum.
Paper is passed over the drum and by charging the paper with opposite charge to the toner, the toner is attracted to the paper and away from the drum.
The paper is heat treated to fuse the toner onto the paper.

Question 21

Optical disk

Answer 21

An optical disk is a generic term for DVDs, CDs and Blu-Ray

It uses laser technology to read/write data.

For read-only optical disks, when data is written it is encoded as of a series of bumps, or pits and lands within the track on the disk. A protective layer is then put over the surface to prevent any corruption of data. The pattern of pits and lands are used to represent data. When the CD is read, the pits and lands are read by the laser which then interprets each as different electrical signals. In turn, the electrical signals can be converted into binary codes.

For writeable optical disks, rather than using pits and lands the disk is coated with a photosensitive dye, which is translucent. When writing to the disk, the laster will alter the state of dye spot that is coated onto the surface making it opaque. The dye reflects a certain amount of light. A write laster alters the density of the dye and a read laser interprets the different densities to create binary patterns which in turn can represent data. Write lasers are higher powered than read lasers.