Character sets: ASCII and Unicode

What you need to be able to do

What a character set is

Text is stored by replacing every character, including spaces and punctuation, with its code. For the word “Cat” in ASCII:

1. Each character has its own unique code in the character set.
2. Each character in the text is replaced by its code (stored in binary).
3. The codes are stored in sequence, in the same order as the characters in the text.

Those three points are exactly how the mark scheme words “explain how the word … is represented”. The examiner's comment (June 2021): the two ideas that were needed are uniqueness and order; “each character has a code” is not precise enough.

ASCII and extended ASCII

Codes for letters and digits run in order, so one given code lets you work out its neighbours:

Unicode

Unicode was designed to give every character in every language its own unique code, plus symbols and emoji, so that text looks the same on every computer in every country.

ASCII vs Unicode

Working with character codes

Mark-scheme focus

“Describe what is meant by a character set” (2 marks)

Credited

• All of the characters / symbols that the computer can use / represent B1
• Each character has a unique number / binary number / hexadecimal number B1

Not credited

• “A character set is a set of characters” (repeats the question)
• “Each character has a code” (not unique)

Why

Two marks need two different points: what the set contains, and the one-to-one link between each character and its code (June 2022 report).

From: 9618/13/M/J/22 Q1(a)(i)

“Explain how the word ‘Clock’ is represented by a character set” (2 marks)

Credited

• Each character has a unique binary code B1
• The binary code for each character is stored in sequence (in the order of the word) B1

Not credited

• “Clock is converted into binary” (how?)
• “Each letter has a code” (unique is missing)
• Listing ASCII codes you can't be sure of

Why

The scheme wants the mechanism: unique code per character, codes stored in the word's order. You never need the actual codes.

From: 9618/13/M/J/24 Q1(d)(ii)

“Give one similarity and two differences between ASCII and Unicode” (3 marks)

Credited

• Similarity: both represent each character using a unique code // both can use 8 bits // ASCII is a subset of Unicode B1
• Difference: Unicode can use up to 32 bits per character, ASCII 7 or 8 B1
• Difference: Unicode can represent a wider range of characters / different languages; ASCII only one language B1

Not credited

• “Unicode is better” / “ASCII is older”
• Two differences that say the same thing (more bits AND more characters can count as one idea if not expanded - give two distinct points)

Why

Each difference must compare both sets on one feature: bits per character, number of characters, languages covered.

From: 9618/11/O/N/22 Q1(c)

Worked examples

Example 1 · ASCII, a word, and Unicode codes (Paper 1, 6 marks) ★ Easy

1 Read it with me what is actually happening?

Three skills in one question: facts about ASCII, how text is stored, and code arithmetic from one given code.

• number of characters\(2^{\text{bits}}\)
• explain how … representedunique code + in sequence
• hexadecimal valueconvert 49
• denary value for '5'count on from '1'

2 What is it asking for? the target, part by part

• (a)128 and 2562
• (b)how HOUSE is stored2
• (c)(i)49 in hex1
• (c)(ii)code of '5'1

3 What do we know? everything the question gives us

• ASCII 7 bits, extended 8
• '1' = 49

• digit characters have consecutive codes

4 Plan the route work backwards from what we want

FindKey ideaAlready have itWork out firstTrap

• (c)(ii)'5'
  • Consecutive'5' is 4 after '1'
  • Trapanswering in binary

49 in hex?

49 = 3 × 16 + 1 → 31.

Which answers are in which base?

(c)(i) hex, (c)(ii) denary: the report saw binary answers to both.

5 Now solve it have a go on paper first, then open the worked solution

1. (a) ASCII: 128 // \(2^7\) B1; extended ASCII: 256 // \(2^8\) B1
2. (b) Each character has its own unique code B1; each character in the word is replaced by its code / the codes are stored in the order in the word B1
3. (c)(i) 49 = 3 × 16 + 1 → 31 B1
4. (c)(ii) 49 + 4 = 53 B1

6 Check it does the answer make sense?

'5' must have a bigger code than '1' and the digits run in order: 53 = 49 + 4 ✓. The report: “a common incorrect answer for extended ASCII was 255”.

Example 2 · Unicode characteristics and an extended ASCII table (Paper 1, 5 marks) ★ Easy

1 Read it with me what is actually happening?

(i) wants two facts about Unicode. (ii) is a table: each row gives you one or two forms of the same code and you fill the gap.

• characteristicsfeatures of Unicode: bits, how many characters, what it covers
• 8-bit binarypad to 8 bits

2 What is it asking for? the target, part by part

• (c)(i)two characteristics2
• (c)(ii)three missing values3

3 What do we know? everything the question gives us

• ! = 33, hex 21
• L = 0100 1100, hex 4C
• ü = 252 = 1111 1100

4 Plan the route work backwards from what we want

FindKey ideaAlready have itWork out firstTrap

• (ii)missing values
  • !33 → 8-bit binary
  • Lbinary → denary
  • übinary → hex

Quick check for 33 in binary?

33 = 32 + 1 → 0010 0001, and hex 21 = 0010 0001 too ✓.

5 Now solve it have a go on paper first, then open the worked solution

1. (i) 8 / 16 / 32 bits per character B1; represents \(2^{16}\) / \(2^{32}\) etc. characters // every language and other characters such as emojis B1
2. (ii) ! : 33 → 0010 0001 B1
3. L: 0100 1100 = 64 + 8 + 4 = 76 B1
4. ü: 1111 1100 → FC B1

6 Check it does the answer make sense?

Each row now has three forms that agree: 4C = 4 × 16 + 12 = 76 ✓, FC = 252 ✓.

Example 3 · Unicode advantages and two code conversions (Paper 1, 4 marks) ★★ Medium

1 Read it with me what is actually happening?

16-bit codes this time. The fastest route for both conversions goes through hex place values (4096, 256, 16, 1).

• advantages … instead of ASCIIcompare: what can Unicode do that ASCII can't
• 16 bitsfour hex digits

2 What is it asking for? the target, part by part

• (b)(i)two advantages2
• (b)(ii)binary → denary1
• (b)(iii)hex → denary1

3 What do we know? everything the question gives us

• 0010 0111 0110 1110(ii)
• 2140 hex(iii)

4 Plan the route work backwards from what we want

FindKey ideaAlready have itWork out firstTrap

• (ii)denary
  • Hex first2 7 6 E
  • Place values2×4096 + 7×256 + 6×16 + 14
• (iii)denary
  • Place values2×4096 + 1×256 + 4×16 + 0

Why go through hex for (ii)?

Adding 16 binary place values is slow and error-prone; four hex digits are quicker.

Is the answer to (ii) bigger than 8192?

Yes: the first hex digit is 2 (2 × 4096 = 8192).

5 Now solve it have a go on paper first, then open the worked solution

1. (i) A wider range of characters can be represented B1 … so characters from more languages can be represented // and symbols such as emojis B1
2. (ii) 0010 0111 0110 1110 = 276E hex = 2 × 4096 + 7 × 256 + 6 × 16 + 14 = 8192 + 1792 + 96 + 14 = 10 094 B1
3. (iii) 2 × 4096 + 1 × 256 + 4 × 16 + 0 = 8192 + 256 + 64 = 8512 B1

6 Check it does the answer make sense?

Both are between 8192 (2000 hex) and 12 288 (3000 hex), as their first hex digit 2 says ✓.

Example 4 · Unicode bits, then one ASCII code in BCD and hex (Paper 1, 3 marks) ★★ Medium

1 Read it with me what is actually happening?

(c) mixes this page with page 1: the code 104 is just a number, so put it into BCD (digit by digit) and into hex.

• BCD value1, 0, 4 → three nibbles
• hexadecimal value104 = ? × 16 + ?

2 What is it asking for? the target, part by part

• (b)bits per Unicode character1
• (c)(i)104 in BCD1
• (c)(ii)104 in hex1

3 What do we know? everything the question gives us

• 'h' = 104

4 Plan the route work backwards from what we want

FindKey ideaAlready have itWork out firstTrap

• (c)(i)BCD
  • Trap0110 1000 is plain binary, not BCD
• (c)(ii)hex
  • ÷ 16104 = 6 × 16 + 8

How many nibbles does BCD of 104 need?

Three: 1, 0 and 4.

5 Now solve it have a go on paper first, then open the worked solution

1. (b) 8 // 16 // 32 (also 64 was accepted) B1
2. (c)(i) 1 → 0001, 0 → 0000, 4 → 0100: 0001 0000 0100 B1
3. (c)(ii) 104 = 6 × 16 + 8 → 68 B1

6 Check it does the answer make sense?

Check (ii) via binary: 104 = 0110 1000 → 6, 8 ✓.

Summary

• Character set: all the characters a computer can represent, each with a unique code.
• Text: each character replaced by its code, codes stored in sequence.
• ASCII 7 bits, 128 characters; extended ASCII 8 bits, 256; Unicode 16/32 bits, 65 536+ characters, all languages, emoji. ASCII is a subset of Unicode.
• Codes are consecutive within letters and digits: count on from a given code; lower case = upper case + 32.

Can I do this?

Quiz

Concept checks

Answer each one in your head first, then open it.

Exam-style practice

New questions, not used above. Do them on paper, then mark your work against the official mark scheme.