A Level Computer Science 9618 Paper 33, May/June 2026

Question paper and mark scheme75 marks11 questions

Solve it on ExamPal

Practise

Untimed. Write on the paper and check as you go.

Start

Exam

Timed, with a clock you set, like sitting the real paper.

Start

Opens full screen.

Download PDF Mark scheme

100%

Opening the question paper

Read as textThe questions and the mark scheme as plain text. Easier to read on a phone.

Question paper

9618/33 · May/June 2026

Download PDF

User-defined data types may be created by programmers when writing software. Explain the reasons why user-defined data types are needed.

[3]

Write the pseudocode statement to declare a non-composite data type, Fibonacci, that holds the following numbers: 0, 1, 2, 3, 5, 8, 13, 21

[2]

Complete the following paragraph using the most appropriate words from the given list. Some words are not used. adjacent Application BitTorrent corresponding Internet layer Link lower protocol route Transport upper The implementation of a .................................................. suite can be viewed as a stack. Each .................................................. of the stack has its own specific functionality and can only communicate with .................................................. layers of the stack. User interaction is associated with the highest level of the stack, which is the .................................................. layer.

[4]

State one example of where serial file organisation could be used.

[1]

Describe how records are organised in serial files.

[2]

Outline the process that is used to find a record in a serial file.

[3]

Describe the characteristics of massively parallel computers.

[3]

State one difference and one similarity between the SIMD and MISD computer architectures. difference .................................................................................................................................. ................................................................................................................................................... ................................................................................................................................................... similarity .................................................................................................................................... ................................................................................................................................................... ...................................................................................................................................................

[2]

Numbers are stored in a computer system using binary floating-point representation with: • 12 bits for the mantissa • 4 bits for the exponent • two’s complement form for both the mantissa and the exponent.

Calculate the normalised binary floating-point representation of –18.5625 in this system. Show your working.

[4]

Explain the problem that would occur if the two binary floating-point numbers representing the denary numbers 91.5 and 123.25 were added together and the result stored in the given system.

[3]

a(i)

The Karnaugh map (K-map) represents a logic circuit with four inputs. Draw loop(s) around appropriate group(s) in the K-map to produce an optimal sum-of-products.

[2]

a(ii)

Write the Boolean expression from your answer to 6(a)(i) as a simplified sum-of-products. Do not carry out any further simplification.

[2]

a(iii)

Write the Boolean expression from your answer to 6(a)(ii) in its simplest form by using Boolean algebra to simplify it.

[1]

Simplify the following expression using De Morgan’s laws and Boolean algebra. Show all the stages in your simplification. $X = \overline{\overline{A + B} + C + \overline{B + \overline{C}}}$

[3]

Outline the process of using a stack to evaluate an expression written using Reverse Polish Notation (RPN).

[3]

Write the following infix expression in RPN. ((18 – 15 + 3) * 5) / 12

[2]

The RPN expression 16 3 * 36 6 / 18 6 - * + is to be evaluated. Show the changing contents of the stack as the RPN expression is evaluated.

[4]

Describe the purpose of quantum cryptography.

[2]

Outline two benefits and two drawbacks of quantum cryptography. benefit 1 .................................................................................................................................... ................................................................................................................................................... ................................................................................................................................................... benefit 2 .................................................................................................................................... ................................................................................................................................................... ................................................................................................................................................... drawback 1 ............................................................................................................................... ................................................................................................................................................... ................................................................................................................................................... drawback 2 ............................................................................................................................... ................................................................................................................................................... ...................................................................................................................................................

[4]

Explain what is meant by regression methods in relation to machine learning.

[3]

The A* algorithm is used to perform searches on a graph. Calculate the best path between the W and E nodes on the following diagram using the A* algorithm. Show your working in the table provided. The first two rows have been done for you.

[5]

A program declares a 1D array globally as PartName[1:1000000] OF STRING, and sorts its contents in ascending alphabetical order. The function PartSearch performs a binary search on the array. The item to be found is passed as a parameter. If it is found, its location in the array is returned, otherwise, –1 is returned.

a(i)

Complete the following table to define the variables used in the function.

[2]

a(ii)

Complete the pseudocode for the function to perform a binary search on PartName using all the variables from the table in 10(a)(i). Some lines may be left blank. FUNCTION PartSearch( ................................................................................................ ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ........................................................................................................................................... ENDFUNCTION

[5]

Describe how the time efficiency of a binary search changes as the number of items in the search list increases. Refer to Big O notation in your answer.

[2]

The table shows assembly language instructions for a processor that has one register, the Accumulator (ACC). label instruction explanation opcode operand LDM #n Load the number n to ACC LDD <address> Load the contents of the location at the given address to ACC LDI <address> The address to be used is at the given address. Load the contents of this second address to ACC ADD <address> Add the contents of the given address to the ACC ADD #n Add the number n to the ACC SUB <address> Subtract the contents of the given address from the ACC SUB #n Subtract the number n from the ACC STO <address> Store the contents of ACC at the given address <label>: <data> Gives a symbolic address <label> to the memory location with the contents <data> # denotes a denary number, e.g. #123 <label> can be used in place of <address> The current contents of memory are: address data 232 120 580 26

Use only the given instruction set to write assembly language code to: • store the denary value 325 in labelled variable A • store the value from location 232 in labelled variable B • subtract the value 50 from the value stored in variable B • store the current value in the accumulator in location 875 • add the value stored in location 580 to the value stored in variable A • subtract the value stored in location 875 from the current value in the accumulator • store the result in variable Final.

[5]

Show the values of the variables A, B and Final from 11(a) in the following table, after the code has been run: label data A: B: Final:

[3]

Mark scheme

9618/33 · May/June 2026

Download PDF

1a · 3 marks

1 mark
The programmer may require a data type for their application that doesn't exist within the programming language
1 mark
… to represent a particular set/group of data / custom data structure
1 mark
… under a single identifier

1b · 2 marks

1 mark
TYPE Fibonacci =
1 mark
(0, 1, 2, 3, 5, 8, 13, 21)

2 · 4 marks

1 mark
protocol
1 mark
layer
1 mark
adjacent
1 mark
Application

3a · 1 mark

1 mark
Batch processing tasks / Creating a transaction file for a batch / Storing bills/payroll / Backup / Appropriate example where data is added chronologically

3b · 2 marks

1 mark
Records are stored in a file one after the other
1 mark
… in chronological order / New records are appended/added at the end of the file

3c · 3 marks

1 mark
Sequential file access is used
1 mark
Each record in the file is searched / checked, one after the other,
1 mark
… until the end of the file is reached / … until the record is found.

4a · 3 marks

1 mark
Massively parallel computers involve the linking together of many processors / computers
1 mark
… to form a more powerful ‘single machine’
1 mark
All the processors work simultaneously on the same problem (with each processor being responsible for its own part of the problem / communication being via interconnected data pathways/networks // shared pools of memory)

4b · 2 marks

1 mark
Difference: SIMD systems all execute the same instruction and MISD systems perform different instructions // SIMD systems work on many data streams/elements simultaneously MISD systems all work on the same data source/element // MISD systems are more complex than SISD systems
1 mark
Similarity: SIMD and MISD are both parallel systems // using multiple processors // both may be pipelined // Neither SIMD nor MISD are well suited for sequential applications such as office productivity software or compilers

5a · 4 marks

1 mark
number converted to binary e.g., positive binary version of 18.5625 = 010010.100100
1 mark
two’s complement version: bits flipped and 1 added = 101101.011100
1 mark
correct mantissa: 1 0 1 1 0 1 0 1 1 1 0 0
1 mark
correct exponent: 0 1 0 1

5b · 3 marks

1 mark
If the two numbers were added together, the value would be 214.75
1 mark
This value exceeds the highest/largest number that can be stored in this system with an exponent of 7, (which is 127.9375)
1 mark
An exponent of 8 would be required to increase the range of possible numbers beyond 128 // Overflow would occur

6a(i) · 2 marks

1 mark
Loop of 4 cells covering column 01 (cells 00,01; 01,01; 11,01; 10,01)
1 mark
Loop of 4 cells covering top and bottom rows of columns 01 and 11 (cells 00,01; 00,11; 10,01; 10,11)

6a(ii) · 2 marks

1 mark
One correct Boolean term ( $\overline{A} . B$ or $B . \overline{D}$ )
1 mark
Both correct Boolean terms connected by an OR sign and no other terms present

6a(iii) · 1 mark

1 mark
$B . (\overline{A} + \overline{D}) // B . (\overline{A . D})$

6b · 3 marks

1 mark
Correct use of De Morgan's laws: $(X =) \overline{A} . \overline{B} . \overline{C} + \overline{B} . \overline{C}$
1 mark
Correct use of Distributive law / Annulment law: $(X =) \overline{B} . \overline{C} . (\overline{A} + 1)$
1 mark
Correct answer: $(X =) \overline{B} . \overline{C} // \overline{B + C}$

7a · 3 marks

1 mark
The RPN expression is read from left to right
1 mark
If the next element is a number it is pushed onto the stack
1 mark
If the next element is an operator, the last two elements are popped from the stack, the operator is applied to them and the result is pushed onto the stack (process repeats until whole expression evaluated)

7b · 2 marks

1 mark
18 15 – 3 +
1 mark
5 * 12 /

7c · 4 marks

1 mark
Ring 1: stack contents for initial operations (push 16, push 3, apply * to get 48)
1 mark
Ring 2: stack contents for middle push operations (push 36, push 6)
1 mark
Ring 3: stack contents for intermediate evaluations (push 18, push 6, apply / and * to get 18, 12, 72)
1 mark
Ring 4: final stack result (120)

8a · 2 marks

1 mark
To provide a more secure method of cryptography/encryption than current methods // To provide a virtually unbreakable method of encryption // To detect eavesdropping.
1 mark
… using the laws of quantum physics rather than mathematical formulae to generate keys // To provide encryption that can be guaranteed for a longer period of time than current methods.

8b · 4 marks

1 mark
Benefit 1: Better security for internet communication // Can detect eavesdropping, enabling the receiver to request another key // Virtually unhackable due to rapidly changing quantum states // Keys are non-duplicable // No-cloning theorem
1 mark
Benefit 2: A second distinct benefit from the list above
1 mark
Drawback 1: Currently can only be used over relatively short distances // Very expensive to implement initially due to costs of installation / hardware / implementation // Requires a dedicated line and specialist hardware // The technology to fully implement quantum cryptography is not currently/readily available // Photon polarization might be impacted in the travelling medium // Terrorists/criminals misuse
1 mark
Drawback 2: A second distinct drawback from the list above

9a · 3 marks

1 mark
A supervised learning technique of machine learning
1 mark
It finds a mathematical function // It plots a best-fit line/curve // calculus is used to find the error gradient (… that maps how independent variables/cause influence a dependent variable/effect between the data)
1 mark
It is used for forecasting trends and predicting outcomes // It will be trained to understand the relationship between different causes and effects allow to find relationships between two sets of data // The goal is to minimise the difference between the predicted values and the actual values.

9b · 5 marks

1 mark
W to N2 and W to N3 (Lines 3 and 4 in table)
1 mark
W ➔ N2 to N5 (Line 5 in table)
1 mark
W ➔ N2 ➔ N5 to N4 (Line 6) / to N6 (Line 7) / to E (Line 8)
1 mark
W ➔ N2 ➔ N5 ➔ N6 to E (Line 9 in table)
1 mark
Final path: W ➔ N2 ➔ N5 ➔ N6 ➔ E

10a(i) · 2 marks

1 mark
First two correct rows in the table
1 mark
Remaining two correct rows in the table

10a(ii) · 5 marks

1 mark
Function header correctly completed
1 mark
Left and right indexes correctly initialised
1 mark
Appropriate loop and escape condition.
1 mark
Setting of the search index (Mid) and conditional statement to find item
1 mark
Conditional statement and changing of left and right indexes if item not found, and return of location if found / -1 if not

10b · 2 marks

1 mark
The time it takes to execute a binary search increases logarithmically as the number of search items increases // The time increases linearly as the number of search items increases exponentially
1 mark
Binary search worst case Big O is O(log n)/ O(log2 n)

11a · 5 marks

1 mark
LDM #325 and STO A
1 mark
LDD 232 and STO B
1 mark
SUB #50 and STO 875 (ignore LDD B)
1 mark
LDD A and ADD 580 // LDD 580 and ADD A
1 mark
SUB 875 and STO Final

11b · 3 marks

1 mark
A: 325
1 mark
B: 120
1 mark
Final: 281

Questions

1
Question 15 marks
Mark scheme ›
2
Question 24 marks
Mark scheme ›
3
Question 36 marks
Mark scheme ›
4
Question 45 marks
Mark scheme ›
5
Question 57 marks
Mark scheme ›
6
Question 68 marks
Mark scheme ›
7
Question 79 marks
Mark scheme ›
8
Question 86 marks
Mark scheme ›
9
Question 98 marks
Mark scheme ›
10
Question 109 marks
Mark scheme ›
11
Question 118 marks
Mark scheme ›