📚 In-depth Past Paper Analysis for Year 12 CIE Computer Science | Year 12 CIE 计算机：历年真题深度解析
In-depth past paper analysis is crucial for mastering CIE AS Computer Science. This article dissects frequently tested concepts, common pitfalls and exam techniques across major topics, using authentic question patterns from recent 9618 papers.
真题深度解析对于攻克 CIE AS 计算机科学至关重要。本文紧扣近年 9618 试卷中的典型题型，深入剖析高频考点、常见错误与应试技巧，帮助考生精准提分。
1. Binary Addition and Overflow Detection | 二进制加法与溢出检测
This topic appears regularly in Paper 2, often asking candidates to add two 8-bit two’s complement numbers and state whether overflow occurs. An overflow happens when the sum falls outside the representable range for the given bit width; in 8-bit two’s complement, that range is -128 to 127.
二进制溢出检测是 Paper 2 的常客，通常要求考生对两个 8 位补码求和并判断是否有溢出。溢出意味着结果超出该位宽的表示范围——8 位补码的范围为 -128 到 127。
Example question: ‘Add the 8-bit two’s complement numbers 01101101 and 01011010. Show your working and explain whether overflow has occurred.’
真题示例：“将 8 位补码数 01101101 和 01011010 相加，写出计算过程，并解释是否发生溢出。”
Key approach: Perform column addition from the least significant bit. 01101101 (decimal 109) plus 01011010 (decimal 90) gives 11000111. Look at the carries into and out of the most significant bit (MSB). Overflow occurs if the carry into the MSB differs from the carry out of the MSB.
解题关键：从最低位开始逐列相加。01101101（十进制 109）加 01011010（十进制 90）得 11000111。观察最高位（符号位）的进位输入和进位输出。若进位输入与进位输出不同，则发生溢出。
Here, carry into MSB = 0 (from bit 6), carry out of MSB = 0. Both are 0, so no overflow. However, careful: The sum 11000111 in two’s complement is -57, which is mathematically incorrect (109+90=199). But because both operands were positive and the result is negative, that indicates an overflow. The carry rule is equivalent: when adding two positives, if result is negative, overflow. In this example, the carries are both 0, so according to the carry rule there is no overflow? Wait, let’s recalculate: 01101101 (109) + 01011010 (90). 0+0=0 cy0; 1+1=0 cy1; 0+1+1=0 cy1; 1+0+1=0 cy1; 1+1+1=1 cy1; 0+0+1=1 cy0; 1+1+0=0 cy1; 0+0+1=1 cy0. Sum = 11000111, carry into MSB (bit 7) from bit 6 is 1, carry out of MSB is 0. So indeed, carry in is 1, carry out 0: different, thus overflow. I earlier misread. Correct: carry into MSB is 1, out is 0, overflow occurs. That matches the sign observation. The answer must state overflow.
此处，进入符号位的进位为 1（来自第6位），符号位进位输出为 0。两者不同，因此发生溢出。若误判则易丢分。答案应明确：发生溢出，因为两个正数相加得到负数。
2. TCP/IP Protocol Stack and Layer Functions | TCP/IP 协议栈与各层功能
CIE often presents a diagram of the TCP/IP four-layer model and asks to name protocols, identify layers for a given task, and explain encapsulation. Candidates must know that the Link layer handles MAC addresses, the Internet layer routes packets using IP, the Transport layer provides end-to-end communication (TCP/UDP), and the Application layer includes HTTP, FTP, SMTP.
真题中常给出 TCP/IP 四层模型图，要求命名协议、为特定任务匹配层级，并解释封装过程。考生须明确：链路层处理 MAC 地址，网际层使用 IP 路由数据包，传输层提供端到端通信（TCP/UDP），应用层包含 HTTP、FTP、SMTP 等。
Sample question: ‘A web browser requests a page from a web server. State the layer at which the browser’s HTTP request is generated, and describe the encapsulation process as data moves down the stack.’
样题：“网页浏览器向网页服务器请求一个页面。说明浏览器发出的 HTTP 请求产生于哪一层，并描述数据在协议栈中向下移动时的封装过程。”
Analysis: HTTP request is built at Application layer. As data descends, Transport layer adds a TCP header (port numbers), forming a segment. Internet layer adds an IP header (source/destination IP), forming a packet. Link layer adds a frame header (MAC addresses) and trailer, producing a frame for transmission.
解析：HTTP 请求在应用层产生。数据向下传递时，传输层添加 TCP 首部（端口号）形成段；网际层添加 IP 首部（源/目的 IP）形成数据包；链路层添加帧首部（MAC 地址）和尾部，封装成帧进行传输。
Common mistake: confusing the order of encapsulation or assigning port numbers to the Internet layer. Always remember: port numbers belong to TCP/UDP at Transport, IP addresses to Internet layer, MAC addresses to Link layer.
常见错误：混淆封装顺序，或将端口号归入网际层。牢记：端口号属于传输层 TCP/UDP，IP 地址属于网际层，MAC 地址属于链路层。
3. Fetch-Decode-Execute Cycle and Register Transfers | 取指-译码-执行周期与寄存器传输
The processing cycle is a core AS topic. A typical exam question provides a diagram of registers (PC, MAR, MDR, CIR, ACC) and asks to describe the steps in order, including data movements on buses.
处理器周期是 AS 级核心内容。常见题型给出寄存器示意图（PC、MAR、MDR、CIR、ACC），要求按顺序描述步骤，并说明总线上的数据传输。
Example: ‘Explain the role of the Program Counter and Memory Address Register during the fetch stage of the cycle. Use the term “address bus” and “data bus”.’
例题：“解释取指阶段中程序计数器（PC）和内存地址寄存器（MAR）的作用，需用到“地址总线”和“数据总线”。”
Model answer in English: The PC holds the address of the next instruction to be fetched. Its content is copied to the MAR via the internal bus. The MAR places this address on the address bus to select the memory location. The control unit issues a read signal, and the instruction is transferred from memory into the MDR via the data bus.
英文标准答案：PC 保存下一条待取指令的地址，其内容通过内部总线复制到 MAR。MAR 将该地址放到地址总线上以选中内存单元，控制单元发出读信号，指令通过数据总线从内存传送到 MDR。
Then the instruction in MDR is copied to the Current Instruction Register (CIR) for decoding. Many candidates omit the step where the PC is automatically incremented after the fetch, which must be stated for full marks.
随后，MDR 中的指令被复制到当前指令寄存器（CIR）进行译码。很多考生遗漏了取指后 PC 自动递增的步骤，这是满分必不可少的细节。
4. Interrupt Handling and the Role of the Interrupt Service Routine | 中断处理与中断服务程序
Questions on interrupts ask about the steps a processor takes when an interrupt signal arrives, including saving the current state on the stack, identifying the interrupt source via the interrupt register, and loading the address of the appropriate ISR into the PC.
中断相关题目会考察处理器收到中断信号后的处理流程，包括将当前状态压入栈、通过中断寄存器识别中断源、将相应 ISR 的地址装入 PC。
Key detail: the contents of the PC and other registers (such as the status register) are saved so that execution can resume later. The interrupt priority is checked if multiple interrupts occur. The interrupt handler is then executed, and upon completion, the saved state is restored.
关键细节：PC 和其他寄存器（如状态寄存器）的内容被保存，以便后续恢复执行。若有多个中断，还需检查中断优先级。执行完中断服务程序后，保存的状态被恢复，继续原程序。
Frequent error: claiming that the ISR address is stored in the MAR during handling. Instead, the processor vectors to the ISR by loading its start address into the PC after saving the context.
常见错误：声称处理过程中 ISR 地址存储于 MAR。实际上，处理器通过将 ISR 起始地址加载至 PC 来跳转，此前先保存上下文。
5. Digital Signatures and Public Key Infrastructure | 数字签名与公钥基础设施
Security and encryption questions require understanding of asymmetric cryptography. A digital signature ensures authenticity and integrity: the sender encrypts a digest (hash) of the message with their private key; the recipient decrypts it with the sender’s public key and compares the hash.
安全与加密题目要求考生理解非对称密码。数字签名用于确保真实性与完整性：发送方用私钥加密消息摘要（哈希），接收方用发送方的公钥解密，并比对哈希值。
Typical question: ‘Explain how a digital certificate provides assurance that a public key belongs to the stated owner.’
典型题目：“解释数字证书如何确保公钥属于声明的主体。”
A digital certificate is issued by a Certificate Authority (CA), which signs it using its own private key. The certificate contains the owner’s identity, their public key, and the CA’s digital signature. The browser uses the CA’s well-known public key to verify the signature, thereby trusting the public key inside.
数字证书由证书颁发机构（CA）签发，CA 使用自己的私钥签名。证书包含持有者身份、其公钥以及 CA 的数字签名。浏览器使用 CA 知名的公钥验证签名，从而信任证书内的公钥。
6. Ethics and Computer Legislation | 伦理与计算机立法
Paper 1 includes questions on ethical issues and legislation such as the Data Protection Act, Computer Misuse Act, and Copyright Designs and Patents Act. Answers must link scenarios to specific legal provisions.
Paper 1 包含伦理与立法题目，如《数据保护法》、《计算机滥用法》、《版权设计和专利法》。作答需结合情景与具体法律条款。
Sample scenario: ‘A company installs keylogging software on employees’ machines without consent. State and explain which law has been broken.’
情景示例：“一家公司未经同意在员工电脑上安装键盘记录软件。指出并解释触犯了哪部法律。”
The Computer Misuse Act prohibits unauthorised access to computer material. Installing keyloggers without permission is unauthorised access with intent to commit further offences, likely falling under Section 2 or Section 3 of the Act.
《计算机滥用法》禁止未经授权访问计算机材料。擅自安装键盘记录器属于意图进一步犯罪的未授权访问，可能触犯该法第 2 或第 3 条。
Candidates should also mention that personalised monitoring may breach the Data Protection Act if personal data is processed unfairly. Clear linkage to the scenario is essential for high marks.
考生还应提及，若对个人信息进行不公平处理，可能违反《数据保护法》。将通用条款与情景紧密结合才能获得高分。
7. Pseudocode for Iteration and Array Traversal | 伪代码中的循环与数组遍历
CIE pseudocode questions in Paper 2 ask to write algorithms using loops to process arrays, such as finding the maximum value, calculating an average, or searching. Standard constructs like FOR ... NEXT, WHILE ... ENDWHILE must be used correctly.
Paper 2 的伪代码题要求用循环处理数组，如查找最大值、计算平均值或搜索。必须正确使用 FOR ... NEXT、WHILE ... ENDWHILE 等结构。
Example: ‘Write pseudocode to count how many elements in array Scores[0:99] are greater than 50, using a FOR loop.’
例题：“用 FOR 循环编写伪代码，统计数组 Scores[0:99] 中大于 50 的元素个数。”
A correct solution: Count ← 0FOR i ← 0 TO 99 IF Scores[i] > 50 THEN Count ← Count + 1 ENDIFNEXT iOUTPUT Count
正确解答如上。许多考生忘记初始化计数器，或者使用错误的循环边界（如 1 TO 100），导致失分。注意 CIE 伪代码中数组索引通常从 0 开始。
8. File Handling with Fixed-Length Records | 定长记录的文件处理
File processing questions ask to read structured data from a file and perform calculations. Understanding how records are defined and accessed is critical, often requiring pseudocode to open a file, read a record, process fields, and close the file.
文件处理题要求从文件读取结构化数据并执行运算。掌握记录的定义与读取方式至关重要，通常需用伪代码打开文件、读取记录、处理字段并关闭文件。
Consider: ‘A file Books.dat stores records of type Book: (ISBN string[13], Title string[30], Price real). Write pseudocode to read all records and output the total price.’
例如：“文件 Books.dat 存储类型为 Book 的记录：（ISBN 字符串[13]，标题 字符串[30]，价格 实数）。编写伪代码读取所有记录并输出总价。”
Answer: OPENFILE "Books.dat" FOR READTotal ← 0WHILE NOT EOF("Books.dat") READFILE "Books.dat", Book Total ← Total + Book.PriceENDWHILECLOSEFILE "Books.dat"OUTPUT Total
解答时需注意：正确使用 OPENFILE、READFILE 读取整个记录，并通过点记法访问字段。文件结束检测（EOF）是循环终止条件，遗漏会导致无限循环或读取错误。
9. SQL Queries with INNER JOIN and Aggregate Functions | SQL 查询中的 INNER JOIN 与聚合函数
Database questions on Paper 2 test SQL SELECT statements involving multiple tables, grouping, and aggregate functions such as COUNT, SUM, AVG. CIE expects ANSI SQL syntax: SELECT ... FROM ... INNER JOIN ... ON ... rather than older comma-style joins.
Paper 2 的数据库题考查多表查询、分组及聚合函数（COUNT、SUM、AVG）。CIE 期望考生使用 ANSI SQL 语法：SELECT ... FROM ... INNER JOIN ... ON ...，而非老式逗号连接。
Sample: ‘Given tables Student(StudentID, Name) and Result(StudentID, Subject, Mark), write SQL to display each student’s name and their average mark.’
样题：“表 Student(StudentID, Name) 和 Result(StudentID, Subject, Mark)，写 SQL 显示每位学生的姓名及其平均分。”
Solution: SELECT Student.Name, AVG(Result.Mark) AS AverageMarkFROM StudentINNER JOIN Result ON Student.StudentID = Result.StudentIDGROUP BY Student.StudentID, Student.NameORDER BY Student.Name;
注意：GROUP BY 需包含非聚合字段，且使用 AS 别名可提高可读性。忘记 ON 条件或误用 HAVING 代替 WHERE 是常见的扣分点。
10. Logic Circuits and Truth Table Construction | 逻辑电路与真值表构建
Boolean logic appears in both papers, requiring truth tables for up to three inputs, and designing circuits from expressions or problem statements. Candidates must be familiar with the notation: A AND B, A OR B, NOT A, and use of XOR in parity checks.
布尔逻辑在两张试卷中均有出现，要求为多达三个输入构建真值表，并根据表达式或情景设计电路。考生需掌握 A AND B、A OR B、NOT A 等符号，以及 XOR 在奇偶校验中的应用。
Example: ‘Construct the truth table for F = (A XOR B) AND NOT(C). Hence draw the logic circuit using AND, XOR and NOT gates.’
示例：“构建 F = (A XOR B) AND NOT(C) 的真值表，并据此画出使用与门、异或门和非门的逻辑电路。”
The truth table has 8 rows. For A,B,C values from 000 to 111, evaluate XOR and then the AND with NOT C. The final column F outputs 1 only when A and B differ and C=0.
真值表包含 8 行。根据 A、B、C 取值 000 到 111，计算 XOR 后再与 NOT C 进行与运算。只有 A 与 B 不同且 C=0 时，F 才为 1。
| A | B | C | A XOR B | NOT C | F |
|---|---|---|---|---|---|
| 0 | 0 | 0 | 0 | 1 | 0 |
| 0 | 0 | 1 | 0 | 0 | 0 |
| 0 | 1 | 0 | 1 | 1 | 1 |
| 0 | 1 | 1 | 1 | 0 | 0 |
| 1 | 0 | 0 | 1 | 1 | 1 |
| 1 | 0 | 1 | 1 | 0 | 0 |
| 1 | 1 | 0 | 0 | 1 | 0 |
| 1 | 1 | 1 | 0 | 0 | 0 |
The corresponding circuit diagram (description): XOR gate output from A and B connects to one input of an AND gate, while a NOT gate provides NOT C to the other input. The AND output is F.
对应电路图（描述）：异或门的输出连接与门的一个输入端，非门提供 NOT C 至另一个输入端，与门输出 F。
11. Data Compression: Lossless vs Lossy | 数据压缩：无损与有损
Paper 1 often contrasts run-length encoding (RLE) with Huffman coding for lossless compression, and JPEG with MP3 for lossy. Candidates must explain how each maintains acceptable quality while reducing file size.
Paper 1 常对比无损压缩中的游程编码（RLE）与哈夫曼编码，以及有损压缩中 JPEG、MP3 的应用。考生需解释如何在减小文件体积的同时保持可接受的质量。
Past paper: ‘Explain why JPEG is suitable for photographs but not for text images. Refer to original data, reconstructed data and file size.’
真题：“解释为何 JPEG 适合照片但不适合文本图像，需提及原始数据、重建数据和文件大小。”
JPEG removes perceptually less important colour variations; the reconstructed image is not identical to the original, but the human eye hardly notices. Text images require sharp edges and exact reproduction, so lossy compression would blur characters. Lossless formats like PNG are needed.
JPEG 抛弃了视觉上不重要的色彩变化，重建图像虽与原图不完全一致，但肉眼难以察觉。文本图像需要清晰边缘和精确重现，有损压缩会导致字符模糊，因此须使用 PNG 等无损格式。
12. Monitoring and Operating System Roles | 监控与操作系统角色
This section covers utility software, multitasking, and memory management. CIE expects candidates to distinguish between scheduler, dispatcher, and memory management functions in the context of a multiprogramming environment.
本节涵盖实用工具软件、多任务处理与内存管理。CIE 期望考生在多重程序设计环境下区分调度程序、分派程序和内存管理功能。
Example: ‘Explain the purpose of a scheduler in a multiprogramming system. Name the three scheduling states and the transition between them.’
例题：“解释多重程序系统中调度程序的目的，并说出三种调度状态及其之间的转换。”
The scheduler decides which process from the ready queue gains the CPU. The states are RUNNING, READY, and BLOCKED. A process moves from RUNNING to READY when its time slice expires, or to BLOCKED when waiting for I/O. It moves from BLOCKED to READY when I/O completes.
调度程序决定就绪队列中的哪个进程获得 CPU。状态包括运行、就绪和阻塞。进程因时间片耗尽从运行转为就绪，或因等待 I/O 而转为阻塞；I/O 完成后从阻塞转为就绪。
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