DSE ICT | Multimedia and data compression — lesson-prep guide for tutors

2026-11-02

Tutor lesson-prep guides · Secondary ICT

Asked for a file size, students write a number that feels about right. This lesson breaks every formula into samples times bits, with units written beside each step.

Objectives — Students can: (1) compute uncompressed image size as width x height x bits per pixel / 8 and explain how colour depth changes it; (2) compute audio size as sample rate x bit depth x channels x seconds / 8; (3) compute a compression ratio from byte counts and choose lossy or lossless for a stated purpose.

01 Prerequisites (3-minute check)

If any of these fails, fix it before the lesson starts.

Skill	Quick check	Pass mark
Bits and bytes	How many bits in a byte, and how many bytes in a megabyte?	Eight bits; one million bytes in decimal
Binary	How many colours can 8 bits represent?	Two to the power eight, which is 256
Sampling	What does a higher sampling rate mean?	More amplitude readings per second, closer to the original sound and a larger file

02 Lesson flow (60 minutes)

Time	Stage	Tutor does	Student does
0-5 min	Opening question	Show a 1920 x 1080 photo beside its file size and ask why it is not one byte per pixel	Guess aloud and state one assumption
5-15 min	Prerequisite check	Three quick items: bytes, powers of two, sampling rate	Written answers, marked live
15-37 min	Three-part script	Teach image size, audio size, then compression	Recap the unit of each symbol in the formula
37-49 min	Question walk-through	Model four questions, writing the unit at every step	Copy and recompute one
49-55 min	Independent practice	Circulate; ask only what the unit of this step is	Do three items and self-mark
55-60 min	Exit check	Collect answers and count misses aloud	Written, no talking

03 Teaching script (read it out)

Scene 1: Part 1: image size and colour depth

A 1920 by 1080 photo has 1920 times 1080 pixels, which is 2,073,600. How many bits per pixel depends on colour depth. True colour at 24 bits uses three bytes per pixel, one each for red, green and blue, giving 6,220,800 bytes, about 6.22 megabytes. At 8-bit indexed colour it is one byte per pixel, so 2,073,600 bytes, about 2.07 megabytes. The same picture drops to one third of the size, at the cost of falling from sixteen million colours to 256. Every question follows one line: pixel count times bits per pixel.

If colour depth rises from 8 bits to 24 bits, how many times larger is the file?

Scene 2: Part 2: audio size and sampling

Sound becomes numbers by sampling the amplitude many times a second and storing each reading in a fixed number of bits. The CD standard is 44,100 Hz, 16 bits, two channels. Bits per second is 44,100 times 16 times 2, which is 1,411,200, so bytes per second is 176,400 after dividing by eight. Three minutes is 180 seconds, giving 31,752,000 bytes, about 31.75 megabytes. Two slips are common: forgetting the channels, and forgetting to divide by eight. Halving the sample rate halves the file but loses the top frequencies.

If stereo becomes mono and nothing else changes, what fraction of the file remains?

Scene 3: Part 3: lossy and lossless compression

Compression comes in two kinds. Lossless, such as ZIP, PNG or FLAC, restores the exact original by coding repeated patterns and typically saves a few tenths. Lossy, such as JPEG or MP3, throws away what eyes and ears notice least: MP3 at 128 kbps gives 2,880,000 bytes for three minutes, a ratio near 11 to 1. The choice is not about which is smaller but about the purpose. Contracts, source code and medical images need lossless; streaming music and web photos can be lossy. Video follows the same logic with three knobs, frame rate, resolution and bit rate, each with a different side effect.

A recording may be used as evidence in court. Would you store MP3 or FLAC, and why?

04 Question types and worked steps

Type	Example	Steps	Answer
Image size	Given resolution and colour depth.	(1) Width x height for pixels; (2) times bits per pixel; (3) divide by 8; (4) divide by 10^6 for MB.	Example: 1920 x 1080 x 24 / 8 = 6,220,800 B, about 6.22 MB.
Audio size	Given rate, depth, channels and duration.	(1) Rate x depth x channels; (2) times seconds; (3) divide by 8; (4) convert to MB.	Example: 44,100 x 16 x 2 x 180 / 8 = 31,752,000 B, about 31.75 MB.
Compression ratio	Given original and compressed size.	(1) Put both in the same unit; (2) original divided by compressed; (3) write as n to 1; (4) judge whether that meets the need.	Example: 31.75 MB against 2.88 MB is about 11 to 1.
Choosing a method	Given a purpose, choose lossy or lossless.	(1) Can the data tolerate loss; (2) what storage or bandwidth limit applies; (3) choose; (4) name the side effect.	Example: web product photos use JPEG; engineering drawings use PNG.

Sanity check (teach this): Reasonableness check: 1920 x 1080 is 2,073,600 pixels; 24 bits is three bytes each, so 6,220,800 bytes, which is 6.22 MB decimal and 5.93 MiB. The 8-bit version is 2,073,600 bytes. Audio at 44,100 x 16 x 2 is 1,411,200 bits per second, so 31,752,000 bytes for 180 s; MP3 at 128 kbps is 2,880,000 bytes, a ratio of 11.02. Network: a 100 MB file on a 100 Mbps link takes 8 s, and an answer of 1 s means the factor of eight between bits and bytes was dropped.

05 Independent practice (3 questions)

Question	Answer	What to watch
Compute the size of an uncompressed 1920 x 1080 image at 24-bit true colour.	2,073,600 pixels times 3 bytes is 6,220,800 bytes, about 6.22 MB.	Check for the three bytes per pixel and the division by eight.
Compute the size of three minutes of CD-quality stereo audio.	44,100 x 16 x 2 x 180 / 8 = 31,752,000 bytes, about 31.75 MB.	Check for the two channels and the 180 seconds.
The same audio is encoded as 128 kbps MP3. Give the ratio.	2,880,000 bytes, so 31,752,000 divided by 2,880,000 is about 11 to 1.	Check that kbps is divided by eight before multiplying by seconds.

06 Common mistakes and what to say

Mistake	Student writes	Say this
Forgetting to divide by eight	Writes 1920 x 1080 x 24 = 49,766,400 MB.	Bits and bytes differ by eight. After multiplying bits, divide by eight to get a file size.
Dropping the channels	Computes 44,100 x 16 and calls it the file.	Stereo is two streams of data, so write the times two for channels, not for bit depth.
Treating colour depth as bytes	Writes that 24 bits equals 24 bytes.	Eight bits make a byte, so 24 bits is three bytes, one each for red, green and blue.
Inverting the ratio	Writes 1 to 11 as the compression ratio.	The ratio is original to compressed, larger number first. Eleven to one means eleven times smaller.

07 Homework (paper, section and question numbers)

Secondary ICT index: finish the three items, then pair binary conversion with the data representation chapter — Exercises 1-3 on this page

08 Exit check (last 5 minutes)

Question	Answer	Decision
How large is an uncompressed 1024 x 768 image at 8 bits?	786,432 bytes, about 0.79 MB	Cannot answer: reteach part 1
Which four quantities appear in the audio size formula?	Sample rate, bit depth, channels, duration	Cannot answer: reteach part 2
What separates lossy from lossless compression?	Whether decompression restores the exact original	Cannot answer: reteach part 3

09 Tutor FAQ

Which year is this for?

S6 ICT multimedia mainly; S5 classes that have covered data representation can start the calculations earlier.

Is a megabyte one million or 1,048,576 bytes?

Follow the paper. With no instruction, use one million bytes and state which convention you used.

Is there a matching paper?

The homework link goes to the secondary ICT index, which pairs with the data representation chapter on binary.

Find a tutor — free Tutor Registration →

A TutorZone lesson-prep guide. Written by the TutorZone team; questions and lesson flow are original and are not taken from any school or publisher material.