📚 Year 11 Cambridge Computer Science: In-Depth Analysis of Past Papers | Year 11 Cambridge 计算机：历年真题深度解析
Cambridge IGCSE Computer Science (0478) past papers are a goldmine for understanding what examiners really look for. This article dissects recurring question patterns, common pitfalls, and effective strategies to turn past paper practice into top grades.
剑桥 IGCSE 计算机科学（0478）历年真题是理解考官评分标准的宝库。本文剖析反复出现的题型、常见失分点以及高效策略，帮助你通过真题训练冲击高分。
1. Understanding the Exam Structure | 理解考试结构
The Cambridge IGCSE Computer Science assessment consists of two compulsory papers. Paper 1, 'Computer Systems', is a 1-hour 45-minute written paper worth 60% of the total marks. It tests theoretical knowledge across the entire syllabus through short-answer and structured questions. Paper 2, 'Algorithms, Programming and Logic', is a 1-hour 45-minute practical-style paper worth 40%. It requires candidates to write pseudocode, interpret flowcharts, complete truth tables and correct given program code.
剑桥 IGCSE 计算机科学考试包含两份必考试卷。试卷一“计算机系统”为1小时45分钟的理论笔试，占总成绩的60%，通过简答题和结构化问题考查全部大纲内容。试卷二“算法、编程与逻辑”为1小时45分钟的实践风格试卷，占40%，要求考生编写伪代码、解读流程图、完成真值表并修正给定的程序代码。
Past papers reveal that Paper 1 often includes a mix of short definitions, comparison tasks, and extended 6–8 mark questions. Paper 2 consistently features a scenario-based programming problem, usually worth 12–15 marks, where you must design an algorithm in pseudocode. Knowing this structure helps you allocate revision time proportionally.
历年真题表明，试卷一通常包含简短定义题、比较题以及6-8分的扩展题。试卷二则必然包含一道基于场景的编程题，通常值12-15分，要求你设计伪代码算法。了解这一结构有助于你合理分配复习时间。
2. Paper 1 Theory: Key Topics That Appear Every Year | 理论卷每年必考的重点主题
Analysis of past papers from 2018 to 2023 shows that certain topics are almost guaranteed to appear. Data representation (binary, hexadecimal, two's complement), logic gates and circuits, input/output devices, and network hardware are perennial favourites. Questions on cybersecurity, encryption, and ethical issues have increased in frequency, reflecting the evolving syllabus emphasis.
对2018至2023年真题的分析显示，某些主题几乎每卷必考。数据表示（二进制、十六进制、二进制补码）、逻辑门与电路、输入/输出设备以及网络硬件是常年热门。随着大纲更新，关于网络安全、加密和伦理问题的考题出现频率有所上升。
A typical Paper 1 question might ask: 'Explain why a designer would choose a solid state drive (SSD) over a hard disk drive (HDD) in a laptop.' The answer expects tiered reasoning: speed, durability, power consumption, and form factor. Simply stating 'faster' only earns one mark; linking the characteristic to the context (portable device, risk of drops) secures full marks.
一道典型的试卷一题目可能会问：“解释为什么设计师会在笔记本电脑中选择固态硬盘而不是机械硬盘。”答案需要分层推理：速度、耐用性、功耗和外形尺寸。仅回答“更快”只能得1分；将特性与场景（便携设备、摔落风险）联系起来才能拿满分。
3. Paper 2 Problem-solving & Programming: What to Expect | 问题解决与编程卷的考查内容
Paper 2 is a practical exam focused on computational thinking. It does not require you to write code in a specific high-level language; instead, you must write pseudocode that follows a clear logical structure. Common tasks include declaring variables, using loops (FOR, WHILE, REPEAT), conditional statements (IF…THEN…ELSE), and array handling. You must also be able to dry-run given code and trace variable values.
试卷二是侧重计算思维的实践考试，不要求使用特定高级语言编写代码，但必须写出结构清晰的伪代码。常见任务包括声明变量、使用循环（FOR、WHILE、REPEAT）、条件语句（IF…THEN…ELSE）以及数组处理。你还需要能够对给定代码进行手工执行并追踪变量值。
One of the most challenging parts is the 'design' question, where you are given a problem statement (e.g., calculate the average of positive numbers entered, stopping when -1 is entered) and you must write a complete pseudocode solution. Past papers show that candidates lose marks by forgetting to validate input, not initialising a counter, or mishandling the termination condition.
最具挑战性的部分之一是“设计”题，你得到一个需求描述（例如，计算输入的正数的平均值，输入-1时停止），并需要写出完整的伪代码解决方案。历年答卷表明，考生常因忘记验证输入、未初始化计数器或没有正确处理终止条件而失分。
4. Data Representation: Common Exam Questions | 数据表示常见考题
Every Paper 1 features at least one data representation question. Binary to denary and hexadecimal conversions, two's complement negative numbers, and binary addition with overflow detection are tested regularly. Increasingly, examiners ask candidates to apply this knowledge to file sizes, colour depths, and sound sampling rates.
每份试卷一都至少有一道数据表示题。二进制与十进制、十六进制转换、二进制补码负数以及带溢出检测的二进制加法是常规考点。考官越来越多地要求考生将这些知识应用于文件大小、色深和声音采样率。
, carry 1. Overflow occurs when the result exceeds the bit width. A classic exam question: 'Add the 8-bit two's complement numbers 01101001 and 01011000. State if overflow occurs and explain why.' The sum is 11000001, which in 8-bit two's complement represents -63 – ive indicates overflow. Explaining the sign bit inconsistency is the mark-winning insight.
对于二进制加法，关键是要记住 1 + 1 = 0，进位 1；1 + 1 + 进位 = 1，进位 1。当结果超出位宽时发生溢出。经典考题：“将8位二进制补码数 01101001 和 01011000 相加。说明是否发生溢出并解释原因。”和为 11000001，在8位二进制补码中表示-63，但两个正数相加得到负数表明溢出。解释符号位的不一致是得分的关键。
5. Communication and Internet Technologies: Typical Pitfalls | 通信与网络技术典型易错点
Questions on network hardware often confuse students. A router forwards data packets between networks using IP addresses, while a switch directs data within a local network using MAC addresses. A hub, now largely obsolete, simply repeats signals to all ports. Past papers show that candidates frequently interchange the functions of routers, switches, and modems, leading to lost marks.
关于网络硬件的问题常常让学生混淆。路由器使用IP地址在网络之间转发数据包，而交换机使用MAC地址在局域网内引导数据。集线器现在基本已被淘汰，它只是向所有端口重复信号。真题显示，考生经常混淆路由器、交换机和调制解调器的功能，导致失分。
Protocol layers (TCP/IP model) are a newer addition to the syllabus. You need to explain the four layers: application (HTTP, FTP, SMTP), transport (TCP port numbers, reliability), internet (IP addressing and routing), and link (MAC, physical transmission). An exam question might provide a scenario of sending an email and ask you to describe the role of each layer. Use the correct terminology: encapsulation, segment, packet, frame.
协议分层（TCP/IP模型）是较新的考纲内容。你需要说明四层：应用层（HTTP、FTP、SMTP）、传输层（TCP端口号、可靠性）、互联网层（IP寻址与路由）和链路层（MAC、物理传输）。一道试题可能给出一个发送电子邮件的场景，要求描述每一层的作用。要使用正确的术语：封装、数据段、数据包、帧。
6. Hardware and Software: Frequently Assessed Concepts | 硬件与软件常考概念
Fetch–decode–execute cycle is a core concept that appears in almost every session. You must be able to describe the role of the program counter (PC), memory address register (MAR), memory data register (MDR), current instruction register (CIR), accumulator, and ALU. Past exam responses show that students who draw a labelled diagram of the CPU and annotate the cycle tend to score higher than those relying solely on text.
取指-译码-执行周期是几乎每场考试都会出现的核心概念。你必须能够描述程序计数器（PC）、内存地址寄存器（MAR）、内存数据寄存器（MDR）、当前指令寄存器（CIR）、累加器和算术逻辑单元（ALU）的作用。历年答卷显示，画出CPU带标签的示意图并注释周期的学生往往比纯粹文字描述的学生得分更高。
Operating system functions are another high-frequency topic. Learn the five key management tasks: memory, process, file, peripheral, and user management. For example, when asked 'How does the OS manage memory?', your answer should cover paging, segmentation, and the role of virtual memory when RAM is full. Use real-world examples from past paper mark schemes, such as 'the OS swaps infrequently used pages to the hard disk'.
操作系统功能是另一个高频主题。要掌握五大管理任务：内存管理、进程管理、文件管理、外设管理和用户管理。例如，当被问到“操作系统如何管理内存？”时，你的答案应涵盖分页、分段以及RAM满时虚拟内存的作用。使用评分标准中的真实例子，如“操作系统将不常用的页面交换到硬盘”。
7. Programming Concepts and Pseudocode: Exam-style Code Walkthrough | 编程概念与伪代码：真题代码解析
Cambridge pseudocode has a specific syntax: variable assignment uses ←, loops use FOR…TO…NEXT or WHILE…DO…ENDWHILE, and arrays are accessed with square brackets. Modifying an incorrectly written algorithm is a common task. For instance, a past paper gave a pseudocode fragment that attempted to find the maximum value in an array of 20 numbers but had a logical error in the loop boundary (checking one element too few).
剑桥伪代码有特定的语法：变量赋值使用 ←，循环使用 FOR…TO…NEXT 或 WHILE…DO…ENDWHILE，数组使用方括号访问。修正错误算法是常见任务。例如，一道真题给出了一个试图查找20个数字数组中最大值的伪代码片段，但循环边界存在逻辑错误（少检查了一个元素）。
Consider this simplified past-paper task: Write a pseudocode algorithm to count and output the number of values in an array Scores[1:30] that are greater than 80. A model answer:
Count ← 0
FOR Index ← 1 TO 30
IF Scores[Index] > 80 THEN
Count ← Count + 1
ENDIF
NEXT Index
OUTPUT Count
Notice the correct loop range, the use of a pre-initialised counter, and the placement of the OUTPUT statement outside the loop. Many candidates erroneously output inside the loop, printing running totals instead of the final count.
考虑这个简化的真题任务：编写伪代码算法，统计数组 Scores[1:30] 中大于80的值的个数并输出。模范答案：
Count ← 0
FOR Index ← 1 TO 30
IF Scores[Index] > 80 THEN
Count ← Count + 1
ENDIF
NEXT Index
OUTPUT Count
注意正确的循环范围、预初始化计数器的使用以及 OUTPUT 语句放在循环外部。许多考生错误地在循环内部输出，打印出运行累加值而非最终计数。
8. Logic Gates and Circuits: Solving Past Questions | 逻辑门与电路：真题解答
Truth tables and logic circuit diagrams are a guaranteed element of Paper 1 and occasionally appear in Paper 2. You must know the symbols and truth tables for AND, OR, NOT, NAND, and NOR gates. More recent papers ask you to complete a truth table for a given logic circuit with three inputs and multiple gates, then write the corresponding Boolean expression.
真值表和逻辑电路图是试卷一的必考内容，偶尔也出现在试卷二中。你必须掌握 AND、OR、NOT、NAND 和 NOR 门的符号与真值表。近年试卷要求你为给定的三输入多级逻辑电路完成真值表，然后写出相应的布尔表达式。
e outputs. For a circuit with inputs A, B, C feeding an AND gate whose output goes to an OR gate with another signal, label the intermediate node. The Boolean expression might be (A AND B) OR (NOT C). In your truth table, systematically evaluate all 2³ = 8 combinations. Examiners often award marks for the intermediate column, so show your working.
常见的考试陷阱是在绘制或求值电路时误解运算顺序。从左到右处理，记下中间输出点。对于一个由输入 A、B、C 进入与门，其输出与另一信号进入或门的电路，标注中间节点。布尔表达式可能是（A AND B）OR（NOT C）。在真值表中，系统地评估所有 2³ = 8 种组合。考官通常会给中间计算列分数，因此务必展示演算过程。
9. Algorithms and Flowcharts: Step-by-step Approach | 算法与流程图：逐步方法
Flowcharts test your ability to visualise logical flow. Start/Stop symbols (oval), process rectangles, decision diamonds, and input/output parallelograms must be drawn clearly. When a past paper asks 'Complete the flowchart to show the given algorithm', missing or incorrect arrows are a frequent cause of lost marks. Always trace your flowchart with sample data to verify the logic.
流程图考查你将逻辑流程可视化的能力。开始/结束符号（椭圆形）、处理矩形、判断菱形和输入/输出平行四边形必须清晰绘制。当真题要求“完成流程图以表示给定算法”时，遗漏或错误的方向箭头是常见的失分原因。务必用样例数据遍历流程图以验证逻辑。
ides the search interval in half. A past paper asked candidates to trace binary search for the value 31 in the array [2, 9, 15, 22, 31, 44, 58, 67]. The first midpoint index (1+8)/2 = 4 (value 22), then discard lower half, new low = 5, high = 8, midpoint = 6 (value 44), discard upper half, low = 5, high = 5, midpoint = 5 (value 31) – found. Write each step in a table to secure all marks.
线性搜索与二分搜索的比较经常出现。要理解二分搜索需要列表已排序，且它反复将搜索区间折半。一道真题要求考生在数组 [2, 9, 15, 22, 31, 44, 58, 67] 中追踪搜索值 31 的二分搜索过程。第一次中位索引 (1+8)/2 = 4（值 22），丢弃低半部分，新低界 = 5，高界 = 8，中位 = 6（值 44），丢弃高半部分，低界 = 5，高界 = 5，中位 = 5（值 31）——找到。用表格记录每一步以获取全部分数。
10. Exam Techniques and Time Management | 考试技巧与时间管理
Past paper practice is only effective if you self-assess using the mark scheme. Highlight command words: 'State' requires a short factual answer; 'Describe' needs a step-by-step account; 'Explain' demands reasoning with because/therefore links. A 4-mark 'Explain' question on why registers are faster than RAM should mention: proximity to CPU, no address bus delay, smaller size, and static RAM technology. Each relevant point with linkage scores a mark.
只有在使用评分标准进行自我评估的情况下，真题练习才真正有效。圈出指令词：'State' 要求简短的事实答案；'Describe' 需要逐步叙述；'Explain' 要求用因为/所以进行推理。一道4分的 'Explain' 题问为什么寄存器比RAM快，应提及：靠近CPU、无地址总线延迟、容量更小以及静态RAM技术。每个有联系的相关知识点得分。
Timing strategy: For Paper 1 (105 minutes for 60 marks), allocate 1.75 minutes per mark. Do not spend more than 15 minutes on an 8-mark question; if stuck, move on and return. In Paper 2, save 20 minutes for the pseudocode design question at the end. Plan the algorithm in rough before writing the final answer, as untidy corrections can confuse the examiner. Regular timed past paper practice builds the stamina and pace required for exam day.
时间策略：试卷一（105分钟完成60分），每题每分钟应完成约1.75分。不要在8分题上花费超过15分钟；卡住时先跳过，回头再做。试卷二中，留出最后20分钟处理伪代码设计题。先在草稿上规划算法再写最终答案，因为凌乱的修改会让考官困惑。规律性的限时真题训练能培养考试所需的耐力和节奏。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图