IGCSE Computer Science (0478): Complete Subject Guide

The principle behind IGCSE Computer Science

Computer science is the study of how problems get solved systematically. Cambridge IGCSE Computer Science (0478) builds on that principle: it teaches you to decompose complex problems into manageable parts, design logical solutions, and express those solutions in code that a machine can execute. Every topic in the syllabus connects back to this central idea. Data representation matters because computers need a way to encode information. Networks matter because isolated machines solve fewer problems than connected ones. Programming matters because it is the language through which human logic becomes machine action.

This subject is not about memorising facts and reproducing them. It is about understanding systems, reasoning through cause and effect, and applying structured thinking under timed conditions. That distinction shapes how you should study and how you should approach the exam itself.

Who should take IGCSE Computer Science

The obvious audience is students who want to pursue software engineering, data science, cybersecurity, or any technology-driven career. IGCSE Computer Science provides the conceptual foundation that A Level Computer Science and university courses build upon directly.

The less obvious audience is equally important. Any student who values logical reasoning and structured problem-solving will benefit from this course. The skills it develops - breaking problems into steps, tracing through processes methodically, thinking in terms of inputs, processes and outputs - transfer to mathematics, engineering, economics, and the sciences. Medical schools increasingly value applicants who can think algorithmically. Law firms recruit people who can reason through conditional logic.

You do not need prior programming experience to take this course. The syllabus teaches programming from first principles. What you do need is a willingness to think precisely. Computer science punishes vague reasoning more quickly than most subjects: a program either runs correctly or it does not, and the gap between the two often comes down to a single logical condition.

The examination structure

IGCSE Computer Science uses two papers, each carrying equal weight. There are no tiers - every candidate sits both papers, and the full grade range from A* to G is available to all.

The absence of tiers is significant. Unlike IGCSE subjects such as Mathematics or the sciences, where Core and Extended papers limit the grade range, Computer Science gives every candidate access to every grade. This means the papers must contain questions that span the full difficulty range, from straightforward recall to complex multi-step problem-solving. Expect the difficulty to escalate as you move through each paper.

Paper 1: Computer Systems

Paper 1 is the theory paper. It tests your understanding of how computers work, how data is represented and stored, how networks operate, and how systems are kept secure. Questions are short-answer and structured, requiring you to explain concepts, perform conversions, apply knowledge to scenarios, and evaluate systems.

The 75 marks across 105 minutes give you roughly 1.4 minutes per mark. That pacing is comfortable for questions requiring short written answers, but tighter for questions that involve binary arithmetic or logic gate analysis. Practising those calculation-heavy topics until they become fluent is the single most effective way to buy yourself time on this paper.

Paper 2: Algorithms, Programming and Logic

Paper 2 is the problem-solving paper. It tests your ability to design algorithms, write and trace code, apply Boolean logic, and work through programming scenarios. You will write pseudocode and program code, complete trace tables, design flowcharts, and debug existing programs.

The same 75 marks and 105 minutes apply, but the cognitive demand is different. Paper 2 questions require you to produce solutions, not just explain concepts. A question might present a problem scenario and ask you to write a complete algorithm from scratch, or give you a faulty program and ask you to identify and correct errors. This is where programming practice outside the classroom pays off directly.

The syllabus: what you need to know

The IGCSE Computer Science syllabus divides into two broad domains that align with the two papers. Understanding this structure helps you see how individual topics connect to each other and to the exam.

Computer Systems (Paper 1 topics)

These topics cover the hardware, software, and theoretical foundations of computing:

• Data representation - binary, hexadecimal and denary number systems; binary addition and overflow; two's complement for negative numbers; character encoding (ASCII and Unicode); image representation (pixels, colour depth, resolution); sound representation (sampling rate, bit depth)
• Computer architecture - the Von Neumann model; the fetch-decode-execute cycle; the role of the CPU, ALU, control unit and registers; primary and secondary storage types; how clock speed, cache size and number of cores affect performance
• Input and output devices - a range of input devices (barcode readers, QR scanners, touchscreens, sensors) and output devices (monitors, printers, speakers); selecting appropriate devices for given scenarios
• Storage devices and media - magnetic, optical and solid-state storage; comparing capacity, speed, portability and durability; choosing storage for specific use cases
• Networks - LAN and WAN; network hardware (routers, switches, access points); client-server and peer-to-peer models; the internet and the World Wide Web; cloud computing; IP addressing and MAC addresses; protocols (TCP/IP, HTTP, HTTPS, FTP, SMTP)
• Cyber security - threats (malware, phishing, social engineering, brute-force attacks, denial of service); prevention methods (firewalls, encryption, authentication, access control); the human factor in security
• Databases - flat-file and relational databases; tables, records, fields and primary keys; SQL queries (SELECT, FROM, WHERE, ORDER BY); data types and validation
• Automated and emerging technologies - robotics, artificial intelligence, sensors in automated systems; ethical, social and environmental implications of computing

Algorithms, Programming and Logic (Paper 2 topics)

These topics cover the problem-solving and coding skills that form the practical heart of the subject:

• Algorithm design and problem-solving - decomposition, abstraction and pattern recognition; pseudocode and flowcharts; standard algorithms (linear search, bubble sort, totalling, counting, finding maximum/minimum); validation and verification
• Programming concepts - variables, constants and data types; assignment, selection (IF/ELSE, CASE) and iteration (FOR, WHILE, REPEAT); arrays (1D and 2D); string handling; file handling (reading from and writing to files); procedures and functions; local and global variables; library routines
• Boolean logic - AND, OR, NOT and NAND gates; truth tables; constructing and simplifying logic expressions; logic circuits

Where the marks concentrate

Not all topics carry equal weight in examinations. Examiners return to certain areas repeatedly because they are central to the discipline and because they differentiate between candidates effectively.

Programming concepts and algorithm design dominate Paper 2. These areas test whether you can actually solve problems, not just describe how solving works in theory. Expect questions that require you to write complete pseudocode solutions, trace through programs step by step, and debug code with subtle logical errors. Strength here is the single strongest predictor of a high Paper 2 mark.

On Paper 1, input and output devices, number systems, and cyber security appear with high regularity. Number system conversions (binary to hexadecimal, denary to binary, binary addition) are examined in almost every session because they are fundamental to understanding how computers process data. Cyber security questions test whether you can reason about threats and defences in realistic scenarios, not just list types of malware.

Boolean logic straddles both papers. Paper 1 may test your understanding of logic gates and truth tables in a theoretical context. Paper 2 will test your ability to use Boolean conditions within programs and to construct logic expressions that solve practical problems.

Computer architecture and databases round out the high-frequency areas. Architecture questions test whether you understand the fetch-decode-execute cycle and can explain how hardware components interact. Database questions test SQL syntax and your ability to design queries that retrieve specific information.

Grade boundaries: what the numbers mean

Cambridge sets grade boundaries after each examination session based on the difficulty of the specific papers sat. Boundaries therefore shift from year to year, and no fixed mark guarantees a particular grade. That said, historical patterns give useful orientation.

For the A* grade, candidates typically need around 85-90% of the total marks across both papers. An A usually requires approximately 75-80%. The B boundary tends to sit around 65-70%, and a C around 55-60%. These are approximate ranges, not guarantees, and a particularly challenging paper can push boundaries lower.

The practical consequence: if you are targeting an A* across two papers totalling 150 marks, you are aiming for roughly 130 marks or above. That leaves little room for error on either paper. Building consistent strength across both papers is more reliable than gambling on a very high score on one to compensate for the other.

Recommended study timeline

The following plan assumes a May/June examination series and a student beginning structured revision in September of the preceding year. Adjust according to when your course finishes teaching new content.

Paper-by-paper strategy

How to approach Paper 1: Computer Systems

Paper 1 tests breadth of knowledge and precision of explanation. The key strategic principles:

1. Read the scenario carefully. Many Paper 1 questions set a real-world context - a school network, a retail database, a smart home system. Your answer must address that specific scenario, not just recite generic textbook definitions. If the question asks why a school should use a star topology, explain the benefits in terms of a school's needs, not in abstract terms.
2. Match your depth to the marks available. A one-mark question needs one clear point. A four-mark question needs four distinct points or a chain of reasoning with four links. Writing a paragraph for a one-mark question wastes time; writing a single sentence for a four-mark question wastes marks.
3. Show your working on calculations. Binary addition, hexadecimal conversion and storage calculations carry method marks. Even if your final answer is wrong, clear working can earn partial credit. Write each step on a separate line.
4. Use technical vocabulary precisely. "The CPU processes data" is too vague. "The ALU performs arithmetic and logical operations on data fetched from RAM via the data bus" demonstrates understanding. Examiners are trained to look for specific technical terms in your answers.

How to approach Paper 2: Algorithms, Programming and Logic

Paper 2 tests your ability to construct solutions. The strategic principles differ because the task differs:

1. Plan before you write. Spend 30-60 seconds thinking through the logic of an algorithm before writing pseudocode. Identify the inputs, the process, and the outputs. Decide which data structures and control structures you need. A minute of planning prevents five minutes of crossing out and rewriting.
2. Trace your solutions mentally. After writing an algorithm, run through it with a simple test case in your head. If the question asks you to sort a list, trace your algorithm with a three-element list to check it produces the correct output. This catches most logical errors before the examiner does.
3. Be precise with pseudocode syntax. Cambridge provides a pseudocode guide in the syllabus. Follow it. Use the correct keywords (DECLARE, INPUT, OUTPUT, IF...THEN...ELSE...ENDIF, FOR...TO...NEXT, WHILE...DO...ENDWHILE). Examiners expect consistency with the published guide, and deviations can cost marks even when your logic is sound.
4. Trace tables must be complete. When asked to trace through a program, show every variable's value at every step. Do not skip iterations because you think the pattern is obvious. The marks are awarded for demonstrating that you can follow the execution path precisely, including the final state of each variable when the program terminates.
5. Boolean logic questions reward systematic working. Complete truth tables row by row, evaluating each sub-expression before combining them. For logic circuit questions, label intermediate outputs at each gate. Rushing through Boolean logic is the fastest way to lose marks on otherwise straightforward questions.

Common mistakes examiners report

Examiner reports identify recurring errors across thousands of scripts. Learning from other candidates' mistakes is one of the most efficient forms of revision.

• Confusing storage units. 1 byte is 8 bits, not 10. 1 kilobyte is 1024 bytes (or 1000 in SI terms - the syllabus uses 1024). Calculation questions that ask for file sizes in different units catch candidates who mix up the conversion factors.
• Writing descriptions instead of explanations. "A firewall monitors network traffic" is a description. "A firewall monitors incoming and outgoing network traffic and blocks packets that do not match predefined security rules, preventing unauthorised access to the network" is an explanation. The difference can be two or three marks on a single question.
• Incomplete pseudocode. Missing ENDIF, NEXT, or ENDWHILE statements. Forgetting to initialise variables before using them in loops. Using assignment where a comparison is needed (= instead of == in conditions). These errors suggest a candidate who thinks about algorithms at a high level but has not practised writing precise code.
• Shallow cyber security answers. "Use a strong password" earns minimal credit. "Use a password with a minimum of 12 characters combining uppercase, lowercase, numbers and symbols, changed every 90 days and not reused across accounts" demonstrates the depth examiners want.
• Mixing up AND/OR in Boolean logic. Candidates often use OR where AND is needed, or vice versa, because they translate from English imprecisely. "A person can vote if they are 18 AND a citizen" uses AND because both conditions must be true. Practise converting English statements into formal Boolean expressions.

Resources that make a difference

The most valuable resource is the syllabus document itself. Cambridge publishes the complete syllabus for 0478 on its website, including the pseudocode guide and the command word definitions. Read it before you begin revision so you know exactly what is examinable and what is not.

Past papers and mark schemes are the second essential resource. Cambridge publishes papers from recent sessions, and working through them under timed conditions is irreplaceable preparation. The mark schemes show you exactly what examiners expect - often with surprising specificity about which words and phrases earn marks.

Examiner reports (published alongside mark schemes) are underused by most candidates. They explain where the cohort performed well and poorly, and they identify the exact misconceptions that cost marks. Reading these reports transforms vague awareness of "common mistakes" into specific knowledge of what to avoid.

For programming practice, any environment that lets you write and run code in Python, Java, or Visual Basic will work. The IGCSE does not prescribe a language for pseudocode answers, but familiarity with a real language makes the abstract concepts concrete. Write programs regularly. Solve problems from past papers in code, then translate your solutions into pseudocode.

Why structured preparation works

IGCSE Computer Science has a clear structure: two papers, defined topics, published assessment criteria. That structure means preparation is not guesswork. You know what will be tested. You know how it will be tested. You know what the examiners are looking for in each answer.

The candidates who perform best are those who treat preparation as a systematic process. They work through the syllabus methodically, practise under exam conditions regularly, analyse their mistakes honestly, and adjust their revision to address weaknesses rather than reinforcing strengths they already have. The subject rewards the same logical, structured approach in revision that it tests in the exam hall.

Start with the foundations, build depth through practice, and refine through past paper analysis. The 150 marks across two papers are earned through understanding and application, not memorisation. That is both the challenge and the opportunity: a student who genuinely understands how computer systems work and can genuinely solve algorithmic problems will find this exam a natural expression of what they know.