Cambridge A‑Level Computer Science 9618 – 1.2 Multimedia & Core Foundations
Learning Objectives
- Explain how data are represented in binary, hexadecimal, BCD and two’s‑complement forms.
- Describe character encodings (ASCII, Unicode) and colour‑depth concepts.
- Identify the main hardware components of a computer system and the role of different memory types.
- Outline the Von Neumann architecture, the fetch‑decode‑execute cycle and the purpose of interrupts.
- Analyse how bitmap and vector images are encoded, calculate their storage requirements and decide which representation is appropriate.
- Understand basic sound sampling, calculate audio file sizes and recognise common audio formats.
- Distinguish between lossless and lossy compression techniques and name the algorithms used in typical multimedia files.
1. Information Representation
1.1 Number Systems
- Binary (base‑2) – digits 0, 1. Used internally by all digital devices.
- Octal (base‑8) – digits 0‑7. Useful for grouping binary in 3‑bit blocks.
- Decimal (base‑10) – digits 0‑9. Human‑friendly.
- Hexadecimal (base‑16) – digits 0‑9, A‑F. Groups binary in 4‑bit nibbles; common in memory addresses.
Quick conversion table (8‑bit values)
| Binary | Hex | Decimal |
|---|
| 0000 0000 | 00 | 0 |
| 0000 1111 | 0F | 15 |
| 0010 1010 | 2A | 42 |
| 1111 1111 | FF | 255 |
1.2 Binary‑Coded Decimal (BCD)
Each decimal digit is stored in a separate 4‑bit nibble. Example: 93₁₀ → 1001 0011₂ (9 = 1001, 3 = 0011). BCD simplifies decimal‑oriented I/O but wastes half the bits compared with pure binary.
1.3 Signed Numbers – Two’s‑Complement
- Range for an n‑bit two’s‑complement integer: –2ⁿ⁻¹ … 2ⁿ⁻¹ – 1.
- Negative value = invert all bits (one’s complement) then add 1.
Example (8‑bit)
+45 → 0010 1101
-45 → invert → 1101 0010, add 1 → 1101 0011 (binary) = –45
1.4 Character Encodings
- ASCII – 7‑bit code (0‑127) covering English letters, digits and control characters.
- Extended ASCII – 8‑bit, adds 128‑255 for additional symbols.
- Unicode (UTF‑8/UTF‑16) – supports > 1 million characters; UTF‑8 is backward compatible with ASCII.
1.5 Colour Depth & Palette
- Colour depth = bits per pixel (bpp). Determines the number of distinct colours:
2^(bpp).
- Palette (indexed colour) stores a table of RGB triples; each pixel holds only an index into that table.
Common colour depths
| bpp | Colours | Typical use |
|---|
| 1 | 2 | Black‑&‑white images |
| 2 | 4 | Early video‑games |
| 4 | 16 | Standard VGA |
| 8 | 256 | GIF, indexed PNG |
| 16 | 65 536 | High‑colour BMP |
| 24 | 16 777 216 | True‑colour photographs |
| 32 | 16 777 216 + alpha | PNG‑32 with transparency |
2. Computer Hardware Fundamentals
2.1 Core Components
- CPU (Central Processing Unit) – executes instructions; contains ALU, registers, control unit.
- Memory hierarchy
- Registers – fastest, inside CPU, a few bytes each.
- Cache – SRAM, speeds up access to frequently used data.
- Main memory (RAM) – volatile; DRAM (dynamic) is common, SRAM is faster but more expensive.
- Secondary storage – HDD, SSD, optical media; non‑volatile.
- Input/Output (I/O) devices – keyboards, mouse, display, speakers, sensors, actuators.
- Bus system – set of parallel wires that transfer data, addresses and control signals (e.g., system bus, address bus, data bus).
2.2 Memory Types
| Memory | Volatile? | Typical technology | Speed (relative) |
|---|
| Register | Yes | SRAM inside CPU | Fastest |
| Cache | Yes | SRAM | Very fast |
| DRAM (main RAM) | Yes | Capacitor‑based cells | Fast |
| ROM (PROM/EPROM/EEPROM) | No | Mask/Flash | Slow |
| SSD/HDD | No | Flash/Nmagnetic | Slowest |
2.3 Embedded Systems (example)
Microcontroller + sensors + actuators → performs a dedicated task (e.g., temperature controller). Emphasise the limited memory and real‑time constraints.
3. Processor Fundamentals
3.1 Von Neumann Architecture
- Single memory stores both data and program instructions.
- CPU fetches an instruction, decodes it, executes it, then repeats – the Fetch‑Decode‑Execute (FDE) cycle.
Typical FDE cycle (register‑transfer notation)
FETCH: MAR ← PC ; address the next instruction
MBR ← MEM[MAR] ; fetch instruction
IR ← MBR
PC ← PC + 1
DECODE: Decode IR (determine opcode, operands)
EXECUTE: Perform operation (ALU, memory access, I/O)
3.2 Interrupts
- Asynchronous signals that pause the current instruction stream.
- CPU saves the current PC (and sometimes status registers) on a stack, jumps to an interrupt service routine (ISR), then restores state.
- Types: hardware (external device), software (system call), timer‑based.
3.3 Example: Simple addition with an interrupt
; Assume accumulator A holds first operand
INT 0x10 ; hardware interrupt to read second operand into B
ADD B ; A ← A + B
4. Communication – Networking Fundamentals
4.1 Network Types & Topologies
| Network | Typical scale | Topology examples |
|---|
| LAN (Local Area Network) | Building / campus | Star, bus, ring |
| WAN (Wide Area Network) | City / globe | Mesh, hybrid |
| PAN (Personal Area Network) | Room / person | Star (Bluetooth) |
4.2 Client‑Server vs Peer‑to‑Peer
- Client‑Server – centralised resources (e.g., web server). Scales well for many clients.
- Peer‑to‑Peer – each node can act as client and server (e.g., file‑sharing).
4.3 IP Addressing
- IPv4: 32‑bit dotted‑decimal (e.g., 192.168.1.25). Supports ~4.3 billion addresses.
- IPv6: 128‑bit hexadecimal groups (e.g., 2001:0db8:85a3::8a2e:0370:7334).
- Subnet mask determines network vs host portion; common masks: /24 (255.255.255.0), /16 (255.255.0.0).
Quick subnet example
Network: 192.168.10.0/24
Valid hosts: 192.168.10.1 – 192.168.10.254
Broadcast: 192.168.10.255
4.4 Core Protocols
- TCP – connection‑oriented, reliable, ordered delivery.
- UDP – connectionless, low‑latency, no guarantee of delivery.
- HTTP / HTTPS – application‑layer protocol for web traffic.
- DNS – translates domain names to IP addresses.
- Ethernet / Wi‑Fi – physical and data‑link layer technologies; CSMA‑CD (wired) and CSMA‑CA (wireless) for media access.
4.5 Example Exam Question (AS)
Given the IPv4 address 172.16.5.130/20, determine the network address and the range of usable host addresses.
Solution
/20 → subnet mask 255.255.240.0
Network address: 172.16.0.0
First host: 172.16.0.1
Last host: 172.16.15.254
Broadcast: 172.16.15.255
5. Multimedia – Bitmap (Raster) Images – Encoding
5.1 Key Concepts
- Pixel – smallest addressable element; identified by (x, y) coordinates.
- Resolution – width × height in pixels.
- Colour depth – bits per pixel (bpp).
- Palette (colour table) – used in indexed‑colour images.
- Bit planes – all bits occupying the same position across pixels; useful for image‑processing.
5.2 Colour‑Encoding Methods
Direct (True‑colour) Encoding
- Each pixel stores its colour directly, usually as three separate channels R, G, B.
- 24‑bit colour = 8 bits per channel → 16 777 216 possible colours.
- 32‑bit colour adds an 8‑bit alpha channel for transparency.
Indexed (Palette‑based) Encoding
- Pixel stores an index (1, 2, 4 or 8 bits) into a palette that holds the actual RGB triples.
- Reduces file size when the image uses a limited set of colours.
5.3 Calculating Bitmap Data Size
Raw size (bits)
\[
\text{bits} = \text{width} \times \text{height} \times \text{bpp}
\]
Convert to bytes (divide by 8) and round up to the nearest whole byte. Many file formats pad each scanline to a 4‑byte boundary – add the padding when required.
Worked Example – 640 × 480, 8‑bpp indexed
- Total pixels = 640 × 480 = 307 200
- Bits = 307 200 × 8 = 2 457 600 bits
- Bytes = 2 457 600 ÷ 8 = 307 200 bytes ≈ 300 KB
5.4 Effect of Changing Resolution or Colour Depth
| Resolution | bpp | File size (KB) | Typical visual quality |
|---|
| 640 × 480 | 8 | 300 | Limited colour, suitable for simple graphics |
| 640 × 480 | 24 | 900 | Rich colour, photographic detail |
| 1280 × 720 | 8 | 1 200 | More detail but still limited colour |
| 1280 × 720 | 24 | 2 764 | High‑definition colour image |
5.5 Data Layout in Memory (row‑major order)
24‑bit RGB (no padding)
Byte 0: Red of pixel (0,0)
Byte 1: Green of pixel (0,0)
Byte 2: Blue of pixel (0,0)
Byte 3: Red of pixel (1,0)
...
8‑bit indexed (no padding)
Byte 0: Palette index of pixel (0,0)
Byte 1: Palette index of pixel (1,0)
...
5.6 Common File Formats & Their Encoding
- BMP – uncompressed RGB or indexed data; optional RLE for 4/8‑bpp.
- GIF – 8‑bpp indexed; uses LZW lossless compression.
- PNG – supports 1‑32 bpp; uses DEFLATE (LZ77 + Huffman) lossless compression.
- JPEG – 24‑bit true‑colour; lossy DCT‑based compression (outside raw bitmap encoding but essential for multimedia).
6. Multimedia – Vector Graphics – Encoding
6.1 What Is a Vector Graphic?
A vector image stores drawing instructions (geometric primitives) rather than colour for every pixel. The file contains coordinates, shapes and styling attributes, making the image resolution‑independent.
6.2 Core Elements
- Objects – line, rectangle, circle, ellipse, polygon, Bézier path.
- Properties – stroke colour, fill colour, line width, opacity, dash pattern.
- Coordinate system – usually a Cartesian grid measured in user units (pixels, points, mm).
Simple SVG Example
<svg width="200" height="100">
<line x1="10" y1="10" x2="190" y2="10"
stroke="blue" stroke-width="2"/>
<circle cx="100" cy="60" r="40" fill="red"/>
</svg>
6.3 File‑Size Considerations
- Size grows with the number of objects and the length of attribute strings, not with image dimensions.
- Very complex drawings (thousands of paths) can become larger than a low‑resolution bitmap.
6.4 When to Use Vector vs. Bitmap
| Situation | Preferred format | Reason |
|---|
| Logos, icons, UI elements | Vector (SVG, EPS) | Scalable, small file, easy colour changes |
| Photographs, textures | Bitmap (JPEG, PNG) | Colour detail cannot be expressed with simple shapes |
| Technical drawings with precise dimensions | Vector (DXF, SVG) | Exact geometry, easy editing |
| Pixel‑art or sprite sheets | Bitmap (PNG, GIF) | Pixel‑level control required |
7. Sound – Basic Encoding
7.1 Fundamental Terms
- Sampling rate (frequency) – samples per second (Hz). Common rates: 44.1 kHz (CD), 48 kHz (DVD), 22.05 kHz (telephone).
- Bit depth – bits per sample; defines dynamic range (e.g., 16‑bit → 96 dB).
- Channels – mono (1) or stereo (2); more channels increase size linearly.
- Pulse‑Code Modulation (PCM) – raw, uncompressed audio representation.
7.2 Calculating Audio File Size (PCM)
\[
\text{bits} = \text{duration (s)} \times \text{sampling rate (samples/s)} \times \text{bit depth} \times \text{channels}
\]
Convert to bytes by dividing by 8.
Example – 10 s of CD‑quality audio
- Samples per channel = 44 100 × 10 = 441 000
- Total bits = 441 000 × 16 × 2 = 14 112 000 bits
- Bytes = 14 112 000 ÷ 8 = 1 764 000 bytes ≈ 1.68 MB
7.3 Common Audio Formats
- WAV – stores raw PCM; large, no compression.
- MP3 – lossy, psychoacoustic model; typical reduction to 1 %–10 % of original size.
- AAC / OGG – modern lossy formats with better quality at similar bitrates.
- FLAC – lossless compression; roughly 50 % size reduction while preserving original data.
8. Compression – Overview
8.1 Lossless vs. Lossy
- Lossless – original data can be perfectly reconstructed (PNG, GIF, FLAC, ZIP). Ideal when any degradation is unacceptable.
- Lossy – some information is permanently discarded to achieve higher compression (JPEG, MP3, AAC). Acceptable when a small loss in quality is tolerable for a large size reduction.
8.2 Typical Algorithms Used in Multimedia
| Algorithm | Type | Typical media | Key idea |
|---|
| RLE (Run‑Length Encoding) |
Lossless |
BMP (4/8‑bpp), simple graphics |
Replace consecutive identical values with a count + value. |
| LZW (Lempel‑Ziv‑Welch) |
Lossless |
GIF, early PNG |
Build a dictionary of repeated patterns. |
| DEFLATE (LZ77 + Huffman) |
Lossless |
PNG, ZIP, gzip |
Sliding‑window duplication + variable‑length codes. |
| DCT‑based (Discrete Cosine Transform) |
Lossy |
JPEG, MPEG‑1/2 video |
Transform blocks to frequency domain, quantise, entropy‑code. |
| Psychoacoustic coding (e.g., MP3) |
Lossy |
MP3, AAC |
Remove audio components masked by louder sounds. |
8.3 Example Exam Question (A‑Level)
Explain why a PNG image of a line drawing is usually smaller than a JPEG of the same picture, even though both have the same pixel dimensions.
Key points for answer
- PNG uses lossless DEFLATE; JPEG uses lossy DCT which introduces artefacts.
- Line drawings contain large uniform areas → RLE/DEFLATE compresses very well.
- JPEG’s block‑based DCT is inefficient for sharp edges; it creates high‑frequency coefficients that increase size.
9. Choosing the Right Representation – Summary Table
| Criteria | Bitmap (Raster) | Vector |
|---|
| Typical content | Photographs, textures, complex colour gradients | Logos, icons, schematics, type‑set text |
| Scalability | Pixelation when enlarged | Resolution‑independent, crisp at any size |
| File‑size trend | Grows with resolution × colour depth | Grows with number of objects & attribute length |
| Editing focus | Pixel‑level colour edits | Shape manipulation, colour swaps |
| Typical formats | JPEG, PNG, GIF, BMP | SVG, EPS, PDF, DXF |
10. Exam‑Style Tips
- Memorise the colour‑depth table and be able to convert bpp → number of colours.
- Practice quick calculations of image and audio file sizes using the provided formulas.
- Know the key differences between lossless and lossy compression; be ready to cite an example of each.
- For networking, be comfortable converting between dotted‑decimal IPv4 and binary, and identifying network/host ranges.
- When asked to choose a representation, reference at least two criteria (e.g., scalability and file size).