Software classification, programming languages and translators – A-level Computer Science exam questions with mark scheme

1Core3 marks

A school's computer technician lists some of the software installed on a student laptop:

a spreadsheet package
a disk defragmenter
a C++ compiler
a photo-editing app
Linux

(a) Complete the classification by stating whether each item is system software or application software. [2]

(b) Explain what is meant by software, distinguishing it from hardware. [1]

(a) Spreadsheet package and photo-editing app identified as application software(1)
(a) Disk defragmenter, C++ compiler and Linux all identified as system software; allow 1 mark (max) if exactly four of the five items are correctly classified across the two points(1)
(b) Software is the (sequences of) instructions / programs that a computer runs, whereas hardware is the physical (electronic/mechanical) components of the computer; both sides needed; accept 'software is intangible, hardware can be touched' only if programs/instructions are mentioned(1)
Answer: (a) Application: spreadsheet, photo-editing app. System: defragmenter (utility), compiler (translator), Linux (OS). (b) Software = programs/instructions; hardware = physical components.
Examiner tip: A compiler is system software (a translator) even though programmers 'use' it directly: classify by what it does for the computer, not by who runs it.

2Core3 marks

System software includes utility programs and libraries.

(a) Give two examples of a task carried out by a utility program. [1]

(b) A programmer writing a mapping app uses a library that provides routines for calculating the distance between two GPS coordinates. Explain two benefits to the programmer of using this library rather than writing the routines herself. [2]

(a) Any two of: backing up files; compressing files; defragmenting a disk; virus/malware scanning; encrypting files; disk formatting/checking; file management. Both needed for the mark; ignore named commercial products(1)
(b) Saves development time / the code is already written so does not need to be written (or designed) again(1)
(b) Library routines are already (thoroughly) tested so are likely to be reliable / fewer errors; accept: written by specialists so may be more efficient/accurate; accept: can be reused in other programs; do not award two marks for the same idea(1)
Answer: (a) e.g. backup, compression, defragmentation, anti-virus. (b) Saves time (no need to write code); pre-tested so more reliable / written by experts.
Examiner tip: 'It is quicker' and 'it saves time' are the same point; the second mark needs a different benefit such as being pre-tested.

3Standard4 marks

A user's laptop is streaming music, running a word processor and sending a long document to a printer, all at the same time.

(a) Explain what is meant by the statement that an operating system hides the complexities of the hardware from the user and from application programs. [2]

(b) Describe two ways in which the operating system manages the laptop's resources in this situation. [2]

(a) The OS provides a virtual machine / an interface (e.g. GUI, system calls / API) so the user or program does not need to know how the hardware works(1)
(a) e.g. an application can save a file / print by calling the OS rather than controlling the disk or printer hardware directly; the OS handles the low-level hardware details (e.g. device drivers); accept any valid illustrative example(1)
(b) One resource management role described in context, from: processor scheduling: allocates processor time to each competing process (e.g. time slices / switching between them) so all appear to run at once; memory management: allocates (and protects) areas of main memory to each process / uses virtual memory; I/O (device) management: manages the printer (e.g. queues/spools the print job, uses a driver) and audio output so processes don't conflict(1)
(b) A second, different resource management role described from the same list (naming a resource without saying what the OS does with it earns no mark)(1)
Answer: (a) OS acts as a virtual machine: programs/user call the OS, which deals with hardware details. (b) Schedules processor time between processes; allocates/protects memory; manages I/O devices such as printer queue.
Examiner tip: Name the resource AND say what the OS does with it among competing processes; 'manages memory' alone is not enough.

Stuck on Software classification, programming languages and translators?

Get it sorted in a free 30-minute session

We work through your weakest questions on this topic on Zoom. I reply the same day.

Free 30-min session on this topic Send me 5 questions a day

4Standard5 marks

A simple processor uses 16-bit machine code instructions. The first 6 bits of each instruction hold the opcode and the remaining 10 bits hold the operand.

(a) State the maximum number of different opcodes that this instruction format allows. [1]

(b) One instruction in a program is stored as the hexadecimal value 1C2F. Write this instruction in binary, and give the opcode and the operand as denary values. [2]

(c) Programs for this processor can be written in assembly language. State the relationship between an assembly language instruction and a machine code instruction. [1]

(d) Give one reason why a programmer would find writing in assembly language easier than writing in machine code. [1]

(a) 64 / 2⁶(1)
(b) 0001 1100 0010 1111(1)
(b) Opcode = 7 and operand = 47; follow through from an incorrect binary conversion if split 6/10 correctly(1)
(c) One-to-one / each assembly language instruction translates into exactly one machine code instruction; reject 'one-to-many'(1)
(d) Uses mnemonics (e.g. ADD, LDR) which are easier to remember/read than binary codes; accept labels/names can be used for memory addresses; accept easier to spot errors / debug(1)
Answer: (a) 64 (b) 0001110000101111; opcode 7, operand 47 (c) one-to-one (d) mnemonics are easier to remember/understand than binary
Examiner tip: Split the binary at the boundary the question gives (6 | 10), not at a nibble boundary: 000111 | 0000101111.

5Standard4 marks

A firm is writing the control software for a small battery-powered sensor that has a very slow processor and only 2 KB of RAM. One developer wants to use assembly language; another wants to use an imperative high-level language such as C.

(a) Explain two advantages of using assembly language for this device. [2]

(b) Explain two advantages of using a high-level language instead. [2]

(a) One advantage explained, from: programmer can optimise code to execute fewer instructions / run faster, which matters on a slow processor; program occupies less memory (smaller machine code), which matters with only 2 KB of RAM; allows direct control/manipulation of hardware (e.g. specific registers, ports) needed by the sensor(1)
(a) A second, different advantage from the list above, explained(1)
(b) One advantage explained, from: easier/quicker to write, read and maintain because it uses English-like statements and structures (e.g. loops, selection, named variables) / one HLL statement translates to many machine code instructions so less code to write; portable/machine-independent: the same source code can be recompiled for a different processor if the sensor's hardware changes; easier to debug / fewer errors; built-in library functions available(1)
(b) A second, different advantage from the list above, explained(1)
Answer: (a) faster, tighter code; less memory; direct hardware control. (b) easier to write/maintain (one statement = many machine instructions); portable to other processors; easier to debug.
Examiner tip: Link each advantage to the scenario (slow processor, 2 KB RAM) to show it is an explanation, not a list.

6Standard3 marks

Describe how a compiler and an interpreter differ in the way they translate and execute a high-level language program. [3]

A compiler translates the whole source code (in one go) before it is run, whereas an interpreter translates and executes it one statement/line at a time(1)
A compiler produces object code / an executable file that can be run again without retranslating; an interpreter produces no object code, so the program must be retranslated every time it runs (and the interpreter must be present)(1)
A compiler reports all errors found (in a list) after translating the whole program; an interpreter stops at the first error it reaches (during execution), and earlier lines will already have run; accept: compiled program runs faster once translated, interpreted program runs slower(1)
Answer: Compiler: translates whole program first, makes object/executable code, reports all errors at end. Interpreter: translates and runs line by line each time, no object code, stops at first error.
Examiner tip: Each mark needs a comparison: describe what BOTH translators do for that aspect.

7Hard4 marks

The diagram shows how a games company distributes one of its games. The source code is translated by Translator X into bytecode, which is then distributed to three different types of device.

(a) Translator X is a type of compiler. State what is meant by bytecode. [1]

(b) Explain how the bytecode is executed on each device. [2]

(c) Explain one reason why the company distributes bytecode rather than compiling the game directly to machine code. [1]

(a) An intermediate language / code (between high-level source code and machine code) produced by compiling the source code; accept instruction set of a virtual machine(1)
(b) Each device has its own virtual machine (written for that processor/platform)(1)
(b) The virtual machine interprets the bytecode / translates it into the machine code of that device's processor (as it runs); accept: compiles it just-in-time(1)
(c) Platform independence / portability: only one version needs to be compiled and distributed and it runs on any device with a virtual machine, rather than compiling a separate version for each processor type (x86, ARM, console); accept: source code is not distributed so it is protected; accept: bytecode can be checked/sandboxed by the VM for security(1)
Answer: (a) intermediate code between source and machine code. (b) each device's VM interprets/translates the bytecode into its own machine code. (c) compile once, run on any platform with a VM.
Examiner tip: Say WHERE the platform-specific step happens: the virtual machine on each device, not Translator X.

8Hard4 marks

A developer times one version of a data-processing program. The source code does not change between runs.

	Translation time (once)	Execution time per run
Compiled	40 s	2 s
Interpreted	0 s (translated as it runs)	9 s

(a) Calculate the smallest number of runs for which the total time taken using the compiler is less than the total time taken using the interpreter. Show your working. [2]

(b) Explain why each run of the interpreted program takes longer than a run of the compiled program. [1]

(c) Despite this, the developer uses the interpreter while writing and testing the program. Give one reason why. [1]

(a) Sets up a valid comparison, e.g. 40 + 2n < 9n / 7n > 40 / n > 5.71; or compares totals at 5 runs (50 s vs 45 s) and 6 runs (52 s vs 54 s)(1)
(a) 6 (runs); reject 5.71 or 5 as final answer(1)
(b) Each statement has to be translated (by the interpreter) every time it is executed / each time the program runs, as well as executed; accept: no object code is kept so translation is repeated(1)
(c) Any one of: no need to wait for (re)compilation after each change; program can be run/tested even if later parts are incomplete or contain errors; stops at the error line so easier to locate errors; can inspect variables / change code part-way through execution(1)
Answer: (a) 40 + 2n < 9n ⇒ n > 5.71 ⇒ 6 runs (52 s vs 54 s). (b) interpreter must translate each statement every run. (c) no waiting for compilation / errors found line by line.
Examiner tip: The answer must be a whole number of runs: 5.71 rounds UP to 6 here, and checking n = 5 and n = 6 confirms it.

9Standard3 marks

A software company sells a spreadsheet application. It distributes the object (executable) code to customers but keeps the source code private.

(a) State the difference between source code and object code. [1]

(b) Explain two reasons why the company distributes object code rather than source code. [2]

(a) Source code is the program as written by the programmer (in a high-level/assembly language); object code is the (machine code) output produced by translating it (e.g. by a compiler), that can be executed (after linking); both needed(1)
(b) One reason explained, from: protects the company's intellectual property, as customers cannot easily read, copy or modify the program code (or see how it works); customers do not need a compiler/interpreter installed; the program runs immediately without being translated, so starts/runs faster; customers cannot introduce errors by editing the code(1)
(b) A second, different reason from the list above, explained(1)
Answer: (a) Source = code as written by programmer; object = translated machine code produced by the compiler. (b) protects IP / code can't be read or copied; customers need no translator and it runs faster.
Examiner tip: 'Object code is secure' earns nothing: say what it protects (the intellectual property) and how (it cannot easily be read or modified).

10Stretch6 marks

A start-up is developing a fitness-tracking system with three parts:

firmware for a wristband with a small, specialised processor, where battery life and speed matter most;
a mobile app that must run on phones with different processor architectures;
short data-analysis scripts that its scientists edit and rerun many times a day.

Discuss the choice of programming language level and translator (assembler, compiler, interpreter, or compilation to bytecode) that would be most suitable for each part. Justify your choices. [6]

Wristband: a compiled HLL (e.g. C) or assembly language translated by an assembler is suitable(1)
Wristband justification: produces machine code that executes directly/fast with small memory footprint; assembly allows code to be optimised and direct hardware control, which helps battery life and speed; no translator needed on the device(1)
Mobile app: compile to bytecode (intermediate code) run by a virtual machine on each phone(1)
Mobile app justification: the same bytecode runs on any processor architecture that has a virtual machine, so only one version has to be produced (portable); source code is not distributed; accept: alternatively compiled separately for each architecture with reason (e.g. faster) but recognising extra work/versions(1)
Scripts: interpreted(1)
Scripts justification: no compile step so changes can be rerun immediately; errors reported at the line where they occur so debugging is quicker; slower execution acceptable for short scripts(1)
Answer: Wristband: compiled C or assembled assembly (fast, compact, hardware control). App: bytecode + VM (portable across architectures). Scripts: interpreter (no recompile, easy debugging; speed less important).
Examiner tip: Each part needs a choice AND a reason tied to the scenario; one generic paragraph on compilers vs interpreters scores poorly.

Stuck on this topic?

Book a free 30-minute session on WhatsApp and we'll find exactly which step is costing you the marks. Or build a timed paper on it.

Free 30-minute session