Edexcel GCSE Computer Science (1CP2) Exam Questions

• B At the end of translating the program (1)
• The only correct answer is B. The correct answer is at the end of translating the program.
• A is not correct because a compiler translates the whole program and does not report an error after each line; reporting after each line is how an interpreter behaves
• C is not correct because a compiler does not stop at the first error; it finishes translating and reports the errors it finds at the end
• D is not correct because syntax errors are found during translation, before the object code is run

Award one mark for:

• The whole program (before it is run) (1)

Allow all of the code/the entire program/the whole program in one go

Do not accept one line/statement at a time or each line as it is executed

Do not accept translating only the lines that contain errors

Award one mark for any of the following up to a maximum of two marks:

• It can control/manipulate the hardware (components/registers/memory locations) directly (1)
• Each instruction maps to a single machine-code instruction, so the behaviour/timing can be controlled precisely (1)
• The code can run very quickly / has a fast execution time (1)
• The code uses less memory/RAM (1)
• The code can be optimised for a particular processor/chipset (1)

Only award one mark where both 'controls the hardware' and 'accesses registers/memory locations' are given, as these are the same point.

Do not accept 'it is closer to the hardware' without saying that it controls or accesses the hardware directly.

Do not accept easier/faster to write/program/learn/understand/debug (these are advantages of high-level languages).

Do not accept faster/quicker without qualification.

Do not accept portable/machine independent.

Do not accept does not need to be translated/compiled (assembly language must be assembled).

Do not accept it is a low-level language / a repeat of the question.

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

• Instructions can manipulate/control the hardware/memory/registers directly (1) so the driver can control the specific device it is written for (1)
• The code is written for one specific processor/architecture (1) so it works with the exact hardware it will control (1)
• The programmer controls exactly how memory/registers are used (1) so the code is efficient and needs little memory, which suits a device that has limited memory (1)
• Each instruction maps to one machine code instruction (1) so the driver code is small enough to fit in the memory of the device (1)

Extra detail, such as naming the device the driver controls, is not needed to earn the marks.

Do not accept 'it is closer to machine code' on its own: it must be developed to show how this helps the driver, e.g. it can control the hardware directly.

Do not accept 'it is faster/quicker' with no link to the hardware/device, e.g. the device responds quickly.

Do not accept 'it is easier to learn', 'it is easier to debug' or 'it takes less time to write'.

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

• The compiler translates the whole program/complete source code in one go (1) whereas the interpreter translates one line/statement at a time (1)
• The compiler translates all of the source code before any of it is run/executed (1) but the interpreter translates each line/statement as it reads it (1)
• The interpreter translates one line/statement at a time (1) and the compiler translates the whole program/source code together (1)

Accept the compiler point and the interpreter point in either order.

Do not accept responses about object code/executable files, when the translated code is executed, how errors are reported, or speed/memory use.

Do not accept 'the interpreter executes/runs each line' on its own, without saying that it translates one line/statement at a time.

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

• The same source code can be used on different processors (1) because a compiler/interpreter for each processor translates it into that processor's machine code/instructions (1)
• The program is machine/processor independent (1) because the code is translated into the machine code/instructions of each processor it is run on (1)
• A different compiler/interpreter is used for each type of processor (1) so the same source code can be translated into each processor's machine code/instructions (1)
• The program can be moved/copied to a computer with a different processor (1) because a compiler/interpreter that produces code for that processor exists (1)

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

Do not accept 'it is portable' without a linked reason about translation for different processors.

Do not accept 'it is easier to use/read/write' on its own without a linked reason about translation/processors.

Do not accept 'it is not written in binary'/'it is not machine code'/'it is not written in assembly language' as the reason without saying that the code is translated into each processor's machine code/instructions.

Do not accept the misconception that the program runs directly on any processor without being translated.

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

• The programmer can control/manipulate the hardware directly (1) because low-level instructions can access registers/memory addresses/hardware features (1)
• The code can be made to work efficiently with a specific device (1) because low-level languages are processor/chipset specific, so the instructions match that hardware (1)
• The code produced is fast to execute (1) because it is optimised for the specific processor/uses few instructions (1)
• The code produced is small (1) so it fits in the limited memory/ROM available in the device (1)

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

Fast execution and small code size are typical advantages of low-level languages; award them only where the linked reason relates to the device.

Do not accept 'it is easier to write/learn/understand' or 'it is faster to write' as a point or as the expansion, because low-level languages are not easier for these.

Do not accept 'it is close to the hardware' or 'it works with hardware' without expansion of what that allows the program to do.

Do not accept 'because it is low-level' as the expansion; the answer must say what the language/hardware allows.

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

• No object code/machine code file is stored/saved (1) so there is no translated program to run on the next occasion (1)
• The translation is only temporary/is lost when the program ends (1) because each line is translated as the program runs (1)
• No executable file is created/saved (1) so the whole program has to be translated again every time it is run (1)
• The interpreter translates each line while the program is running (1) so nothing is kept that could be run without the interpreter (1)

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

Do not accept 'it is interpreted' or 'it is not compiled' without saying that no object code/executable is stored/saved.

Do not accept 'the source code is high-level so it must be translated' or 'the source code has to be turned into machine code' on its own, as this is true of any translation.

Do not accept answers about the source code being deleted.

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

• The compiler translates the whole program into machine code before it is run (1) so no translation is needed while the program is executing (1)
• The machine code/object code is already produced before execution (1) so the CPU can carry out the instructions directly without translating them (1)
• The compiled program does not need an interpreter at run time (1) so no time is spent translating each line before it is executed (1)

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

A statement that translation happens before execution, with no linked reason for faster running, earns 1 mark only.

Do not accept 'faster to run' or 'quicker' on its own without a reason linked to translation before/during execution.

Do not accept 'compiled code is optimised' or 'compilation takes less time' without linking this to translation before execution/at run time.

Do not accept answers about the compiler being installed or available.

The question asks why a compiled program usually runs faster, so answers must refer to when translation happens, not just general performance claims.

Indicative content

The output of each type of translator

• A compiler translates the whole program in one go and produces object code.
• The object code is saved as an executable file, which is already in machine code so the processor can run it directly.
• An interpreter does not produce object code, so no translated file exists for the company to save and sell.
• An interpreter translates one line at a time and executes it immediately, but that translation is not kept.

Whether the program has to be translated each time it is run

• Compiled code is translated once, so the executable can be run as many times as the customer likes without being translated again.
• If the source code is edited, the whole program has to be compiled again to produce a new executable.
• An interpreted program is translated again every time it is run, so a customer who runs the program often waits for translation at every run.
• An executable runs faster than interpreted code because it is already in machine code.

What the customer needs on their computer to run the program

• A customer who installs the executable needs no compiler on their computer, so the program will run straight away.
• If the company supplied the source code instead, the customer would need an interpreter installed, and it would have to be an interpreter for their type of computer.
• A customer given only the executable does not have the source code, so cannot change the program.

Conclusion

• A compiler is the better choice for this company, because the executable can be sold and the customer runs it without it being translated again and without being given the source code.
• Using an interpreter would mean supplying the source code to customers and translating the program on every run.

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 cover all three bullets for this company's program, including translating once rather than on every run, and reach a conclusion.

Do not accept 'a compiler fixes errors' (a compiler reports syntax errors at the end of translation, it does not fix them) or 'a compiler is easier to use' with no link to output, translation or the customer. Do not accept answers about translating to assembly language.