Worked example · Computer Science · Year 10

Calculating file sizes for a school podcast

LNF: Numeracy → Understanding the number system helps us to represent and compare relationships between numbers and quantities → Calculation

What this is

A complete, secure Year 10 answer, annotated so you can see which success criterion each part meets. It is also the answer key. Headline answers: thumbnail 900 KB; logo 64 KB; cover art 5.61 MB; jingle 2.52 MB; interview 30.28 MB; episode 40.94 MB (does not fit); compressed episode 10.02 MB (fits); audio reduced by 87.5%. Answers that differ by 0.01 because of rounding earlier are fine. The most common mistakes are forgetting to change 12 minutes into seconds and dividing by 1,024 only once.

Step 1: units warm-up

QuestionAnswer
Bits in 3 bytes3 × 8 = 24
Bytes in 2 KB2 × 1,024 = 2,048
KB in 5 MB5 × 1,024 = 5,120
4,096 bytes in KB4,096 ÷ 1,024 = 4
Bytes in 1 MB1,024 × 1,024 = 1,048,576

Going to a bigger unit means dividing; going to a smaller unit means multiplying. Evidences criterion 2.

Step 2: images

ImageWorking in bitsBytesKBMB
Thumbnail640 × 480 × 24 = 7,372,800921,6009000.88
Logo256 × 256 × 8 = 524,28865,536640.06
Cover art1,400 × 1,400 × 24 = 47,040,0005,880,0005,742.195.61

(b) Thumbnail: KB (900 KB is a whole number; 0.88 MB is less than 1). Logo: KB (64 KB). Cover art: MB (5.61 MB is easier to read than 5,742.19 KB). A sensible unit gives a number that is not tiny and not huge.

Each row shows the multiplication, then divides by 8, then by 1,024, then by 1,024 again. Evidences criteria 1, 2 and 5.

Step 3: sound

SoundWorking in bitsBytesKBMB
Jingle44,100 × 16 × 15 × 2 = 21,168,0002,646,0002,583.982.52
Interview22,050 × 16 × 720 × 1 = 254,016,00031,752,00031,007.8130.28

(b) 12 × 60 = 720 seconds. The sample rate is in samples per second, so the length must be in seconds too. Using 12 would make the answer 60 times too small.

Units are matched before multiplying. Evidences criteria 1 and 2.

Step 4: will it fit?

(a) Audio: 2 × 2.5234 + 30.2811 = 35.33 MB. Episode: 35.33 + 5.61 = 40.94 MB. This is more than 25 MB, so it does not fit.

(b) Compressed audio: 35.33 ÷ 8 = 4.42 MB. Episode: 4.42 + 5.61 = 10.02 MB. This is less than 25 MB, so it fits, with about 15 MB to spare.

(c) The audio is now 1/8 of its size, so it has lost 7/8. 7 ÷ 8 = 0.875 = 87.5% reduction. Check: (35.33 − 4.42) ÷ 35.33 × 100 = 87.5%.

The pupil compresses only the audio, as the stimulus says, and checks the percentage by two methods. Evidences criteria 3 and 5.

Step 5: what if?

If the interview was recorded…New sizeHow I know
at 44,100 Hz60.56 MBSample rate doubles, so size doubles: 30.28 × 2
in stereo60.56 MB2 channels instead of 1, so size doubles
at 8 bits15.14 MBBit depth halves, so size halves: 30.28 ÷ 2
for 6 minutes15.14 MBLength halves, so size halves

(b) Harri: 31,752,000 ÷ 1,000,000 = 31.75 MB. Mine is 31,752,000 ÷ 1,048,576 = 30.28 MB. He divides by 1,000,000 but I divide by 1,024 × 1,024 = 1,048,576, which is bigger, so my answer is smaller. The difference is about 5%.

Every setting is multiplied in the formula, so each change scales the size in proportion. The pupil explains the 1,000 and 1,024 difference with numbers. Evidences criteria 2 and 4.

If you finish early: answers

Why this response is secure