dcf.cymru stimulus sheet · Computer Science · Year 10

Lleisiau’r Cwm (Voices of the Valley): the digital archive brief

The project, the people and the files on this sheet are invented. The formulas, the storage units and the file types are real and are the ones in the GCSE Computer Science specification.

1. The project

Lleisiau’r Cwm is a volunteer group in the Cynon Valley. Local families have lent them old photographs, postcards, chapel records and recorded interviews with people who worked in the pits, the shops and the chapels. The volunteers want to:

keep a master copy of every item at the best quality they can, so nothing is lost if the original fades or breaks;
put web copies on a public website so anyone can look and listen;
share everything through one cloud storage account. The group can only afford the 15 GB plan.

Gareth, the chair, writes: “We have no idea how big these files will be. Will it all fit? Should we squash the files, and will that spoil them?”

2. Storage units

Computer Science uses base 2 units: each step up is 1,024 times bigger, not 1,000.

Unit	Size
bit	a single 0 or 1
nibble	4 bits
byte	8 bits
kilobyte (KB)	1,024 bytes
megabyte (MB)	1,024 KB
gigabyte (GB)	1,024 MB
terabyte (TB)	1,024 GB
petabyte (PB)	1,024 TB

To go down a unit, multiply by 1,024 (or by 8 from bytes to bits). To go up a unit, divide by 1,024 (or by 8 from bits to bytes).

3. The formulas

Image size in bits = width in pixels × height in pixels × colour depth in bits Sound size in bits = sample rate in Hz × bit depth × length in seconds

Colour depth is the number of bits used for each pixel. 1 bit gives 2 colours (black or white). 8 bits gives 256 shades. 24 bits gives about 16.7 million colours.
Sample rate is how many times a second the sound is measured. 48,000 Hz means 48,000 samples every second.
Bit depth (for sound) is the number of bits used to store each sample.
All the recordings on this sheet are mono (one channel), so the formula above is all you need. Ignore file headers and other extra data.
After working out the bits, divide by 8 for bytes, then by 1,024 for KB, then by 1,024 again for MB.

4. The items to digitise

Photographs and documents, scanned

Code	Item	Scan size (pixels)	Colour depth
P1	Colliery band, 1950s, black and white print	2,048 × 1,536	8 bits (256 shades of grey)
P2	Colour postcard of Aberdare Park, 1970s	1,600 × 1,024	24 bits
P3	Handwritten page of chapel minutes, A4	2,480 × 3,508	1 bit (black or white only)
P4	Small preview picture for the website	320 × 240	24 bits

Sound recordings, all mono

Code	Item	Sample rate	Bit depth	Length
A1	Interview with a retired miner	48,000 Hz	16 bits	40 minutes
A2	A choir singing one hymn	44,100 Hz	16 bits	3 minutes
A3	An old answerphone message	8,000 Hz	8 bits	10 minutes

5. What compression did

A volunteer saved P2 and A1 in different formats. The original is the uncompressed size from the formula.

File	Format	Type of compression	Original	Compressed
P2 postcard	PNG	lossless	4,800 KB	2,400 KB
P2 postcard	JPEG	lossy	4,800 KB	480 KB
A1 interview	FLAC	lossless	225,000 KB	140,625 KB
A1 interview	MP3	lossy	225,000 KB	37,500 KB

Lossless compression finds patterns and stores them more efficiently. When the file is opened, every bit of the original comes back exactly.
Lossy compression throws away detail that people are less likely to notice, such as tiny colour changes or very high and very quiet sounds. That detail is gone for good.
Compression ratio = original size : compressed size, written as something : 1. For example, 900 KB squashed to 300 KB is 900 ÷ 300 = 3 : 1.
Reduction in file size = (original − compressed) ÷ original × 100. For the same example: 600 ÷ 900 × 100 = 66.7%.

6. The whole collection

1,200 photographs, each the same size as P2 (4,800 KB uncompressed).
60 interviews, each the same as A1 (225,000 KB uncompressed).
The cloud account holds 15 GB. Everything, master copies and web copies, has to go in it.
The group also owns a 256 GB external drive, kept in the chapel vestry.