Edexcel GCSE Computer Science (1CP2) Exam Questions

Award up to two marks for a linked description such as:

• The same source code can be run on different types/makes of processor (1) with few or no changes being made to the program (1)
• A compiler/interpreter is available for each type of processor (1) to translate the source code into the machine code for that processor (1)
• The statements are not tied to one processor's instruction set (1) so the program is portable between different computers (1)

Do not accept 'it is translated by a compiler' on its own.

Do not accept 'the program does not need to be translated' or 'it is easier to use'.

Award up to two marks for a linked explanation such as:

• The interpreter translates and executes each line in turn (1) so it cannot continue past a line it cannot translate (1)
• The interpreter finds the error when it reaches that line (1) so it stops and displays an error message before later lines are executed (1)
• Each line is translated just before it is executed (1) so a syntax error in that line prevents that translation/execution from happening (1)
• The interpreter does not translate the whole program first (1) so it only discovers the error when that line is reached during execution (1)

For both marks, the expansion must follow/associate with the statement.

Do not accept 'the program crashes' without saying that the interpreter cannot translate/execute the line.

Do not accept answers about how a compiler reports errors.

Indicative content

How each translator deals with a syntax error

• An interpreter stops as soon as it finds a syntax error and displays an error message.
• With an interpreter, the lines before the error have already been executed, so the student can see how far the program got before it stopped.
• A compiler reports errors at the end of translation, so the student gets a list of all the syntax errors in the program in one go.

How much of the program is translated at a time

• An interpreter translates one line at a time and executes each line immediately.
• A compiler translates the whole program in one go before any of it is run.
• A compiler produces object code, so after every change the whole program must be compiled again before the student can run it.
• An interpreter produces no object code, so the student does not have to wait for the whole program to be translated before it runs.

How this affects the student's testing

• An interpreter suits this student because the program can be run straight away after each small change.
• The interpreter stops at the first error found, so the student fixes that error and runs the program again, but may only find the next error on the next run.
• Reporting every error at the end of compilation means the student can fix several errors at once, but cannot run the program until the whole program compiles.
• The student should use an interpreter while testing, and compile the finished program once it works.

Levels

• Level 0 (0 marks): No rewardable content.
• Level 1 (1–2 marks): Basic, independent points are made, showing elements of understanding of key concepts/principles of computer science. (AO1) The discussion will contain basic information with little linkage between points made or application to the context. (AO2)
• Level 2 (3–4 marks): Demonstrates adequate understanding of key concepts/principles of computer science. (AO1) The discussion shows some linkages and lines of reasoning with some structure and application to the context. (AO2)
• Level 3 (5–6 marks): Demonstrates comprehensive understanding of key concepts/principles of computer science to support the discussion being presented. (AO1) The discussion is well developed, with sustained lines of reasoning that are coherent and logically structured, and which clearly apply to the context. (AO2)

Level 3 responses must relate the way each translator works to the student's testing, not just describe the two translators.

Do not accept answers about run-time or logic errors where no comparison of the translators is made.

Indicative content

Control of the hardware

• Low-level languages can read and write the contents of registers and memory addresses directly, so the firmware can take the values from the distance sensors without calling a library routine.
• Assembly language instructions can read from and write to specific input/output ports on the vacuum cleaner's processor, so the motors can be switched on and off directly.
• The programmer decides exactly which processor operations are carried out and in which order, which matters when the vacuum must reverse away from a step or an obstacle.
• A low-level language puts nothing between the program and the hardware, so the programmer can control the hardware of the vacuum cleaner precisely.

Efficiency of the code

• Assembly language code can be hand-optimised instruction by instruction, so the code that checks a sensor is as short as possible.
• The programmer can choose which registers hold values that are used again and again, which cuts down the number of memory accesses the firmware needs.
• A hand-written low-level program can be made compact, so it is more likely to fit in the small amount of memory the vacuum cleaner has.
• Because each instruction does one defined job, the time taken to react to a sensor reading is predictable, which suits firmware that must not overrun.

Link between the language and the processor

• Low-level languages are processor-specific, so the instructions used are those of the exact processor in the vacuum cleaner rather than general statements that need translating.
• Each assembly language instruction maps to one machine code instruction, so the programmer knows precisely which processor operation will be carried out by each line of code.
• An assembler turns each mnemonic into its machine code equivalent, so the machine code running on the vacuum cleaner's processor is exactly what the programmer wrote.
• Code written for a different processor would have to be rewritten, which is acceptable here because the vacuum cleaner always uses the same processor.

Levels

• Level 0 (0 marks): No rewardable content.
• Level 1 (1–2 marks): Basic, independent points are made, showing elements of understanding of key concepts/principles of computer science. (AO1) The discussion will contain basic information with little linkage between points made or application to the context. (AO2)
• Level 2 (3–4 marks): Demonstrates adequate understanding of key concepts/principles of computer science. (AO1) The discussion shows some linkages and lines of reasoning with some structure and application to the context. (AO2)
• Level 3 (5–6 marks): Demonstrates comprehensive understanding of key concepts/principles of computer science to support the discussion being presented. (AO1) The discussion is well developed, with sustained lines of reasoning that are coherent and logically structured, and which clearly apply to the context. (AO2)

Within Level 1, award 1 mark for a single basic point and 2 marks for two or more independent basic points; Level 2 requires the points to be linked to the vacuum cleaner; Level 3 requires both control of the hardware and efficiency of the code, and a few concise, linked sentences can reach Level 3, so extra length is not required.

Accept registers, memory addresses, input/output ports, machine code and mnemonics as ways of describing access to the hardware, and accept 'a low-level language is faster' as a basic point on its own. Do not award 'binary is a low-level language' (machine code is written in binary, but binary is not a language), or characteristics of high-level languages such as being easy to read or portable, as characteristics of a low-level language.