📚 Year 11 CAIE Computer Science: High-Frequency Exam Topics and Common Mistakes | Year 11 CAIE 计算机：高频考点与易错题分析
Mastering CAIE IGCSE Computer Science (0478) requires more than memorising facts; you must be able to apply concepts, avoid common pitfalls, and think like a computer scientist. This revision guide highlights the topics that appear year after year on Paper 1 and Paper 2, and unpacks the mistakes that cost students the highest marks. Whether you struggle with two’s complement, trace tables, or network layers, the paired English–Chinese explanations below will reinforce your understanding and help you walk into the exam hall with confidence.
掌握CAIE IGCSE计算机科学（0478）不仅仅需要记忆知识点，更需要能够灵活应用概念、避开常见陷阱，并以计算机科学家的思维来解题。这份复习指南聚焦于试卷一和试卷二中几乎每年必考的高频主题，并深入剖析那些让考生白白丢分的典型错误。无论你在二进制补码、跟踪表还是网络层次上感到困惑，下面英中对照的讲解都会巩固你的理解，帮助你有信心地走进考场。
1. Data Representation and Number Systems | 数据表示与数制
Conversions between binary, denary, and hexadecimal form the backbone of Paper 1. A frequent error is misreading the direction of conversion—always double-check whether the question asks for binary to hex or hex to binary. For two’s complement, the most significant bit (MSB) is the sign bit: 1 indicates a negative number. To decode a negative value, flip all bits and add 1, then treat the result as a positive magnitude and prefix a minus sign. Do not forget that 8-bit two’s complement can represent integers from -128 to +127; many students incorrectly state the lower bound as -127.
二进制、十进制和十六进制之间的转换是试卷一的基础。考生常犯的一个错误是看错转换方向——务必确认题目要求的是二进制转十六进制，还是十六进制转二进制。在二进制补码中，最高有效位是符号位：1表示负数。要解读一个负数的值，需要将所有位取反后加1，把结果视为正数大小，再在前面加上负号。别忘了8位补码可以表示的整数范围是-128到+127；许多学生错误地把下限写成-127。
Overflow occurs when the result of an addition exceeds the representable range. A classic exam trap: adding two positive numbers in two’s complement and obtaining a negative result, or adding two negatives and getting a positive. Students often miss overflow because they stop at the binary sum without checking the sign bits of the operands and the result. Always compare the carry into the MSB with the carry out; if they differ, overflow has happened.
当加法运算的结果超出了可表示范围时，就会发生溢出。一个经典的考试陷阱是：两个正数用补码相加却得到了一个负数结果，或者两个负数相加得到了正数。学生经常因为只算出了二进制和，却没有检查操作数和结果的符号位而遗漏溢出判断。要始终比较进入最高位的进位和从最高位出来的进位：如果它们不同，就发生了溢出。
2. Logic Gates and Truth Tables | 逻辑门与真值表
CAIE uses the standard ANSI/IEEE symbols for logic gates. Mixing up the distinctive shapes—especially AND (bullet-shaped) and OR (shield-shaped)—is a simple but costly mistake. When completing a truth table, proceed column by column from inputs towards the final output, and label intermediate outputs if the circuit has multiple gates. A common slip is misreading a NAND gate as AND followed by NOT, but then applying De Morgan’s laws incorrectly.
CAIE采用标准的ANSI/IEEE逻辑门符号。混淆不同的形状——尤其是AND（子弹形）和OR（盾形）是一个简单却代价高昂的错误。在补全真值表时，要一列一列地从输入向最终输出推进，如果电路包含多个门，要对中间输出做标注。另一个常见失误是把NAND门看作“先AND再NOT”，却又错误地运用了德摩根定律。
For an XOR gate, remember that the output is 1 only when the two inputs are different. Students frequently extend this logic incorrectly and claim that a 3-input XOR behaves the same way; in fact, a multi-input XOR outputs 1 when an odd number of inputs are 1. Stick to the 2-input definition given in the syllabus unless told otherwise.
对于异或门，要记住只有当两个输入不同时输出才为1。学生经常错误地推广这个逻辑，认为三输入异或门也是如此；实际上，多输入异或门在输入中有奇数个1时才输出1。除了题目另有说明，遵守教学大纲中给出的二输入定义即可。
3. System Architecture and the Fetch–Execute Cycle | 系统架构与取指执行周期
The von Neumann architecture stores both data and instructions in the same memory. In the fetch–execute cycle, the Program Counter (PC) holds the address of the next instruction. The instruction is copied to the Current Instruction Register (CIR) via the Memory Data Register (MDR). A typical blunder is to say the instructions are decoded in the MDR; decoding occurs in the Control Unit (CU) after the CIR receives the instruction. Also, remember that the address bus is unidirectional (from processor to memory), while the data bus is bidirectional.
冯·诺依曼架构将数据和指令存放在同一个存储器中。在取指执行周期里，程序计数器保存着下一条指令的地址。指令经过内存数据寄存器被复制到当前指令寄存器。一个典型的错误是说指令在MDR中被解码；实际上是CIR收到指令后，控制单元才进行解码。同时要记住，地址总线是单向的（从处理器到内存），而数据总线是双向的。
Exam questions may ask you to describe the role of each register after a specific clock pulse. Write bullet points that clearly separate the fetch phase, decode phase, and execute phase. Avoid vague statements like “the instruction is processed”; instead, state exactly what happens to the PC, MAR, MDR, CIR, and Accumulator (ACC). Practice with a diagram of the registers and buses, tracing arrows for each step.
考题可能会要求你描述某个特定时钟脉冲后各个寄存器的作用。要写出清晰的要点，把取指阶段、解码阶段和执行阶段分开。避免使用“指令被处理”这样含糊的表述；而要准确说明PC、MAR、MDR、CIR和累加器各自发生了什么。练习时可以对照寄存器和总线图，为每一步画出箭头。
4. Input and Output Devices | 输入与输出设备
Questions on sensors and actuators often ask for a real-world application. For example, a temperature sensor sends an analogue signal that is converted by an ADC before a microprocessor can read it. A common error is stating that the sensor itself outputs digital data—most sensors produce analogue signals. Similarly, students confuse OCR (Optical Character Recognition) with OMR (Optical Mark Recognition): OCR reads printed or handwritten characters, whereas OMR detects the position of marks on a pre-printed form, such as a lottery ticket or multiple-choice answer sheet.
关于传感器和执行器的题目经常要求举出实际应用。例如，温度传感器发出模拟信号，该信号经模数转换器转换后才能被微处理器读取。常见的错误是声称传感器本身输出数字数据——实际上大多数传感器产生的是模拟信号。同样，学生常混淆光学字符识别和光学标记识别：OCR读取印刷或手写的字符，而OMR检测预印表格上标记的位置，比如彩票或选择题答题卡。
When asked to justify the choice of an output device, link its features to the context: laser printers for high-volume, high-quality text; inkjet printers for colour photos at low volume; 3D printers for prototyping; projectors for large audiences. A pitfall is recommending a device because “it is fast” without explaining why speed is essential in that scenario.
当题目要求论证选择某种输出设备的原因时，要将其特性与使用场景相联系：激光打印机适用于大批量、高质量的文本输出；喷墨打印机适合小批量的彩色照片；3D打印机用于原型制作；投影仪适合大范围观众。一个陷阱是仅仅因为“它速度快”就推荐某设备，却没有解释为什么在该场景中速度至关重要。
5. Storage Technologies | 存储技术
Distinguish between primary storage (RAM and ROM) and secondary storage (HDD, SSD, optical discs). RAM is volatile, stores data and instructions currently in use, and loses its content when power is off. ROM is non-volatile and typically holds the boot program (BIOS). Students regularly lose marks by saying that ROM stores the operating system; the OS is loaded from secondary storage into RAM at boot.
要区分主存储器（RAM和ROM）与二级存储器（HDD、SSD、光盘）。RAM是易失性的，存储当前正在使用的数据和指令，断电后内容丢失。ROM是非易失性的，通常存储引导程序（BIOS）。学生常常因为说ROM存储操作系统而失分；实际上操作系统在启动时从二级存储器加载到RAM中。
Compare magnetic, optical, and solid-state storage in terms of capacity, speed, portability, durability, and cost. A common misconception is that solid-state drives (SSDs) have moving parts because they are described as “drives”; in reality, they use flash memory with no moving parts, which makes them more shock-resistant and power-efficient than HDDs. However, SSDs have a limited number of write cycles, a detail that examiners love to test.
在比较磁性、光学和固态存储时，要从容量、速度、便携性、耐用性和成本等方面入手。一个常见的误解是，因为被称为“驱动器”就认为固态硬盘有移动部件；实际上SSD使用闪存芯片，没有移动部件，因此比HDD更抗震、更省电。但SSD的写入次数有限，这个细节考官非常喜欢考查。
6. Networks and the Internet | 网络与因特网
When explaining the role of network hardware, be precise: a switch directs data to the specific device on a LAN using MAC addresses, while a router connects different networks and forwards packets based on IP addresses. A hub, which is largely obsolete, simply broadcasts data to all ports. Mixing up the functions of a router and a switch is one of the most frequent errors on this topic.
在解释网络硬件的作用时要准确：交换机利用MAC地址把数据引导到局域网中的特定设备，而路由器连接不同的网络并根据IP地址转发数据包。集线器基本已过时，它只是把数据广播到所有端口。混淆路由器和交换机的功能是这个话题上最常犯的错误之一。
IP addresses are logical and can change, while MAC addresses are physical, unique, and hardcoded into the NIC. The TCP/IP stack has four layers (application, transport, internet, link). Students often place HTTP in the transport layer—HTTP belongs to the application layer. When drawing or describing packet structure, remember that a packet contains the source and destination IP addresses in the header, a sequence number, and a payload; the trailer holds an error-checking component like a checksum.
IP地址是逻辑地址且是可变的，而MAC地址是物理地址，具有唯一性并且固化在网卡中。TCP/IP协议栈有四层（应用层、传输层、网络层、链路层）。学生经常把HTTP放在传输层——HTTP属于应用层。在绘制或描述数据包结构时，要记住包头中含有源IP地址和目的IP地址、序号及有效载荷；包尾则含有用于错误检测的校验和等成分。
7. Cyber Security Threats and Prevention | 网络安全威胁与防范
Be ready to define and compare malware types: virus (attaches to a host file, needs human action), worm (self-replicates across networks without user intervention), Trojan horse (masquerades as legitimate software), spyware (monitors user activity), and ransomware (encrypts files and demands payment). For each, explain the potential impact and at least one specific prevention method, such as using anti-malware software, firewalls, or not opening unknown email attachments.
要准备好定义和比较各类恶意软件：病毒（附着在宿主文件上，需要人为操作）、蠕虫（无需用户干预即可在网络中自我复制）、特洛伊木马（伪装成合法软件）、间谍软件（监视用户活动）以及勒索软件（加密文件并索要赎金）。对每一种都要说明潜在影响和至少一种具体的防范方法，例如使用反恶意软件、防火墙，或不打开未知的电子邮件附件。
A common mistake is to say that a firewall removes viruses. A firewall monitors and controls incoming and outgoing network traffic based on predetermined rules; it does not scan for or delete viruses. Brute-force attacks attempt to guess passwords by trial and error; mitigation includes account lockout policies and CAPTCHA. Social engineering, like phishing, exploits human psychology rather than technical weaknesses—always link “phishing” with emails that appear to be from trusted sources.
一个常见错误是声称防火墙可以清除病毒。防火墙根据预设规则监控进出网络的流量，它不会扫描或删除病毒。暴力攻击通过反复试错来猜测密码，缓解措施包括账户锁定策略和验证码。社会工程学攻击，如网络钓鱼，利用的是人类心理而非技术漏洞——一定要把“钓鱼攻击”和伪装来自可信来源的电子邮件联系起来。
8. Algorithm Design and Trace Tables | 算法设计与跟踪表
Trace tables are used to test the logic of an algorithm by recording the values of variables at each step. The most frequent errors are: not initialising new variables in the table, not following the exact order of execution (especially inside loops), and forgetting to update the loop counter. Always write down the line number or statement being executed, and only change one variable per row unless multiple changes occur in one step.
跟踪表通过记录每一步变量值来测试算法逻辑。最常犯的错误包括：没有在表中初始化新变量、没有遵循准确的执行顺序（尤其在循环内部），以及忘记更新循环计数器。一定要记下正在执行的行号或语句，并且每行只更改变一个变量，除非某一步发生了多个变化。
Standard algorithms such as linear search, binary search, bubble sort, counting occurrences, and finding maximum/minimum appear repeatedly. For binary search, the array must be sorted. Many candidates write a linear search when asked for a more efficient method on large sorted datasets. In pseudocode, use meaningful variable names, correctly apply loops (FOR … TO … NEXT, WHILE … ENDWHILE), and always close any conditional or iterative block with ENDIF, ENDWHILE, or NEXT. Don’t use a single ‘=’ for assignment; the syllabus expects the left arrow ←.
标准算法如线性搜索、二分搜索、冒泡排序、计数以及寻找最大/最小值反复出现。二分搜索要求数组必须先排好序。当题目要求在一个大型有序数据集上采用更高效的方法时，许多考生却写了线性搜索。在伪代码中，要使用有意义的变量名，正确运用循环结构（FOR … TO … NEXT，WHILE … ENDWHILE），并且总是用ENDIF、ENDWHILE或NEXT来关闭条件或循环块。不要用单个等号“=”赋值；教学大纲要求使用左箭头←。
9. Pseudocode and Programming Concepts | 伪代码与编程概念
CAIE pseudocode is a high-level language that is not case-sensitive but follows strict structural rules. Input is obtained with INPUT, output with OUTPUT, and assignment uses ←. IF statements must have THEN and ENDIF. In CASE-OF-OTHERWISE-ENDCASE structures, remember to include an OTHERWISE clause to handle unexpected input. A typical lose-mark scenario: writing OUTPUT “Hello”, x without a semicolon or comma; the syllabus uses commas to separate items in OUTPUT, though colons are also accepted if used consistently.
CAIE伪代码是一种高级语言，不区分大小写，但遵循严格的结构规则。用INPUT获取输入，OUTPUT输出结果，赋值使用←。IF语句必须有THEN和ENDIF。在使用CASE-OF-OTHERWISE-ENDCASE结构时，记得包含OTHERWISE子句来处理意外输入。一个常见的丢分场景是：写了OUTPUT “Hello”, x却没有用逗号分隔项；大纲规定用逗号分隔OUTPUT中的各项，冒号若全文一致也可接受。
Data types matter: integer, real (float), char, string, Boolean. Declare variables implicitly or with explicit comments; the loop variable in a FOR loop is usually an integer. Array indexing starts at 1, not 0, unless stated otherwise. This is a major trap for students who are used to Java or Python. Read the question carefully: some exam papers state that arrays are 0-indexed, but by default CAIE assumes 1-based. When using 2D arrays, the syntax is ArrayName[row, column].
数据类型很重要：整型、实型（浮点型）、字符型、字符串型、布尔型。变量可以隐式声明或通过注释显式说明；FOR循环中的循环变量通常是整型。数组索引默认从1开始，而非0，除非题目另有说明。这对于习惯Java或Python的学生来说是一个大陷阱。认真读题：有些试卷会说明数组是0索引，但CAIE默认采用从1开始的索引。使用二维数组时，语法为ArrayName[row, column]。
10. Databases and SQL | 数据库与SQL
On Paper 1, you need to define key database concepts: table, record, field, primary key, foreign key, and data redundancy. A primary key uniquely identifies each record in a table; a foreign key is a primary key from another table used to create relationships. A common slip is to say that a foreign key must be unique—it does not, because one table can have multiple records referencing the same parent record. Normalisation reduces data redundancy and avoids update anomalies, but you do not need to perform normalisation steps; just understand the rationale.
在试卷一中，你需要定义关键的数据库概念：表、记录、字段、主键、外键和数据冗余。主键唯一标识表中的每一条记录；外键是另一个表的主键，用来建立关系。一个常见的口误是说外键必须唯一——它不是唯一的，因为一个表中可以有多条记录引用同一条父记录。规范化减少了数据冗余并避免更新异常，但你不必执行规范化步骤，只需理解其基本原理。
Simple SQL queries appear on Paper 2. The basic syntax is SELECT field1, field2 FROM TableName WHERE condition ORDER BY field3 ASC/DESC. When writing queries, enclose text values in single quotes, e.g. WHERE City = ‘London’. Use wildcards with LIKE: % represents zero or more characters, _ represents a single character. Remember that SQL keywords are case-insensitive, but table and field names may be case-sensitive depending on the question. Avoid adding unnecessary semicolons unless the question specifies; examiners may accept them, but clarity is key.
试卷二会出现简单的SQL查询。基本语法是SELECT field1, field2 FROM TableName WHERE condition ORDER BY field3 ASC/DESC。书写查询时，用单引号括起文本值，例如WHERE City = ‘London’。在LIKE中使用通配符：%表示零个或多个字符，_表示单个字符。记住SQL关键字不区分大小写，但表名和字段名可能会根据题目要求区分大小写。除非题目明确要求，不要画蛇添足地添加分号；考官可能接受，但清晰才是关键。
11. Data Integrity and Validation | 数据完整性与验证
Validation checks whether data is sensible and reasonable before processing; verification checks that the data has been entered correctly. The differences are tested almost every series. Examples of validation: range check, type check, length check, presence check, format check, and check digit. Verification methods include double entry and visual check. A recurring mistake is labelling a parity check as a verification method when it is actually an error-detecting method used during data transmission.
验证是在数据处理前检查数据是否合理、有意义；校验则是确认数据输入是否正确。两者的区别几乎每轮考试都会涉及。验证的例子包括：范围检查、类型检查、长度检查、存在性检查、格式检查和校验位。校验方法包括双重录入和目视检查。反复出现的一个错误是把奇偶校验标记为校验方法，而它实际上是数据传输中使用的错误检测方法。
Check digits are calculated using an algorithm (e.g., ISBN-13, UPC). Students often fail to correctly apply the modulus operator when performing the calculation. Practice a few examples: for ISBN-13, each digit is weighted alternately by 1 and 3, the sum is divided by 10, and the remainder is subtracted from 10 to give the check digit. If the remainder is 0, the check digit is 0. Always show all steps in your working—method marks are available even if the final digit is wrong.
校验位是用特定算法计算出来的（如ISBN-13、UPC）。学生在计算时经常不能正确使用取模运算符。要练习几个例子：对于ISBN-13，每一位数字交替乘以1和3的权重，求总和后除以10，用10减去余数就得到校验位。如果余数为0，则校验位为0。一定要写出所有解题步骤——即使最终数字错了，也有方法分可拿。
12. Ethical, Legal, and Environmental Issues | 道德、法律与环境问题
This section bridges computer science and society. Be prepared to discuss the Data Protection Act (or equivalent), Computer Misuse Act, Copyright and Patents, and the digital divide. For each, state who is affected, what the law prohibits, and the consequences of breaking it. For instance, unauthorised access to computer material is an offence under the Computer Misuse Act; posting copyrighted software on a file-sharing site violates copyright law.
这一部分连接了计算机科学与社会。要准备好讨论数据保护法、计算机滥用法、版权与专利权，以及数字鸿沟。对每一项都要说明谁受到影响、法律禁止什么行为以及违反的后果。例如，未经授权访问计算机资料属于《计算机滥用法》规定的违法行为；将受版权保护的软件发布在文件共享网站上则违反了版权法。
Environmental topics include the energy consumption of data centres, e-waste disposal, and the carbon footprint of manufacturing devices. Students often write generic statements like “recycling is good”. Instead, link specific actions: reclaiming precious metals from circuit boards reduces the need for mining; cloud computing can optimise server utilization and reduce idle power. Also, remember to discuss the ethical implications of artificial intelligence, such as bias in algorithms and job displacement—these are increasingly appearing in exam papers.
环境话题包括数据中心的能源消耗、电子垃圾的处理，以及制造设备所产生的碳足迹。学生常常写出“回收利用是件好事”这样笼统的句子。正确的做法是联系具体的行动：从电路板中回收贵金属可以减少采矿需求；云计算可以优化服务器利用率并降低空转功耗。此外，还要记住讨论人工智能的伦理影响，比如算法偏见和工作岗位被替代——这些正越来越多地出现在试卷中。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导