A computer stores bits, not letters, emojis or accented names. To store text, a system needs an agreed mapping between each permitted character and a numeric value. That mapping is a character set. When text is saved, the characters are replaced by code values; when it is displayed, software uses the same convention to turn those values back into symbols.

If the writer and reader use different conventions, the bit pattern can be decoded as the wrong character. This is why text systems specify an encoding: agreement is more important than memorising isolated character-code values.

Early systems needed a compact, shared set of English letters, digits, punctuation and control characters. ASCII met that need, but it was not designed to represent every writing system. Modern software often needs names, currency symbols, mathematical notation and scripts from many languages, so Unicode provides a much larger common repertoire.

Unicode does not mean that every character always takes the same number of bytes. UTF-8 uses one byte for the ASCII-compatible part of Unicode and more bytes for many other characters. Therefore, only calculate a fixed text size when the question states the character set or the number of bytes/bits per character.

A character set maps symbols to numeric code points; an encoding maps those code points to bytes. Unicode is not one fixed number of bytes per character, so storage must use the named encoding, such as UTF-8. A bitmap stores colour values for pixels, giving uncompressed size approximately width × height × bits per pixel; metadata and compression change actual file size. Vectors store shapes mathematically and scale without pixelation, but are not ideal for photographic detail.

For a fixed-width character representation, each stored character needs the same number of bits. Count every stored character: letters, digits, punctuation, spaces and line breaks if the question tells you they are present. Then multiply by the stated number of bits per character. Convert to bytes only after calculating the total number of bits.

A bitmap image stores a rectangular grid of pixels. Each pixel has a stored colour value. The pixel data must be read in the correct order and with the correct number of bits per pixel; otherwise the same sequence of bits could produce the wrong colours or wrong image shape.

A bitmap stores colour values at fixed positions; it does not store an object called “a circle” or “a person”. Enlarging a bitmap must spread its existing pixels across more display positions unless extra detail is supplied from elsewhere. That is why enlarged low-resolution photographs can look blocky or pixelated.

Image resolution is the number of pixels used to represent an image, often expressed as width × height. A 1200 × 800 bitmap contains 960 000 pixels. If the image records genuine source detail, increasing the pixel count can make edges and small features appear smoother. It also creates more pixel values to store.

Colour depth (also called bit depth) is the number of bits assigned to each pixel. With $b$ bits per pixel, there are up to $2^b$ distinct bit patterns and therefore up to $2^b$ available colours or shades. Greater colour depth can reduce visible banding in gradients and represent subtler colour variation, but it increases the storage required for every pixel.

For an uncompressed bitmap, first calculate the number of pixels, then multiply by the colour depth. This gives bits. Divide by eight for bytes. Include a file header, row padding or compression only when the question gives information about them; otherwise state that the result is an estimate for the pixel data.

A vector graphic does not store a colour value for every position in a pixel grid. It stores a drawing list: an ordered collection of drawing objects and their properties. The software renders those instructions to create the visible image at the required size.

When a vector graphic is enlarged, the renderer recalculates the geometry of its objects for the new size. It can therefore remain sharp at different scales. This is especially useful for logos, maps, icons, technical plans and simple illustrations. A highly detailed photograph would need an enormous and impractical number of objects, so a bitmap is normally more appropriate for that kind of content.

The best format is determined by the task, not by a slogan such as “vectors are smaller” or “bitmaps are better quality”. Identify what the user needs to preserve or do: photographic detail, clean scaling, frequent editing of shapes, a particular output size, or a limited download budget. Then connect that need to the stored representation.

Many products use both formats. A company may maintain its logo as vector artwork, export a bitmap version at the exact dimensions required for a web banner, and place that banner beside photographs that are naturally bitmaps. The correct answer can therefore be a mixed solution when the scenario contains different kinds of visual content.

Marking note: the calculation needs units at each stage. The explanation needs both the limitation of ASCII and the practical reason Unicode fits the system. Saying only “Unicode is bigger” is incomplete.

Check: at 8 bits per pixel, each pixel needs exactly one byte, so the final byte total equals the pixel count. This is a useful sense check, not a replacement for showing the method.

Marking note: two choices alone are not enough. Each justification must connect the required property of the asset to how the representation stores and renders information. Do not claim that vector graphics are always smaller; complexity affects vector-file size too.

A museum guide must display visitors’ names and exhibit labels in several scripts, show high-detail photographs of artefacts, and use the museum logo on a phone screen, a large information display and printed material. This is one product with three different representation problems, so one file type is not automatically suitable for everything.

Textbook-quality answers do more than describe the normal case. They state the assumption behind a method, identify what fails when the assumption is false, and give the relevant consequence. This matters particularly when questions mix file storage, display quality and format choice.

Extended responses often start with a short technical definition and then require its consequence. The strongest answers make the causal chain explicit. For example, “a bitmap is a grid of pixel values” is a definition. To explain file size, continue: “therefore increasing width or height creates more pixel values to store, and increasing colour depth uses more bits for each pixel.”

Command words: describe asks what is stored or what happens; explain asks for the cause or consequence; justify requires a reason tied to the scenario. A correct statement without its consequence is often only partial credit.

Marking rationale: the numerical method earns marks for pixel count, bit count and byte conversion. The text answer earns credit for linking Unicode to the multilingual requirement. The vector answer earns credit for linking scalable drawing objects to the required large output, not merely for naming the format.