R093 TA4: File Formats and Properties

(a) JPG, because lossy compression gives a small file at a quality that is acceptable for a photograph on screen, so the page loads quickly. (b) SVG, because it stores the logo as mathematical paths, so it stays perfectly sharp at any size from a favicon to a banner and supports a transparent background.

Lossy compression reduces file size by permanently throwing away data that the eye or ear is least likely to notice, so the original can never be recovered and quality drops further with each re-save. JPG is a lossy format, and so is MP3. Lossless compression reduces size by storing the same information more efficiently, so the file can be decompressed back to an exact copy of the original. PNG and FLAC are lossless, as is ZIP. The choice is a trade: lossy gives much smaller files for delivery, lossless keeps the master intact for editing and archiving.

The print version must be CMYK, because a printing press lays down cyan, magenta, yellow and black ink and an RGB file has to be converted at some point; converting it at the start means the designer sees the colours that will actually be printed rather than being surprised by dull greens. It needs to be around 300 ppi at final size so that detail and type edges are sharp on paper, and it should be supplied as a press ready PDF with bleed and crop marks so that trimming does not leave white edges. The web version must be RGB, because screens emit red, green and blue light, and a CMYK file either will not display correctly or will be converted by the browser with unpredictable results. It only needs enough pixels for the size it is displayed at, so a far lower resolution is appropriate, and it should be exported as a compressed JPG or WEBP of a few hundred kilobytes. The properties differ because the two destinations have opposite priorities: print is a fixed, one chance process where quality cannot be recovered, while the web is a bandwidth limited medium where a slow page loses the audience before the poster is seen.

Each deliverable goes to a different destination, and the destination decides the format rather than any general idea of quality. The leaflet is going to a commercial printer, so the artwork should be built in CMYK at around 300 ppi at final trimmed size and delivered as a press ready PDF with bleed, crop marks and fonts embedded or converted to outlines. That specification exists because print is a single irreversible process: a low resolution image or an unexpected colour conversion cannot be corrected after the run, and a missing font substituted by the printer's system would change the design. The social video should be exported as MP4 using the H.264 codec, because that combination plays on effectively every phone, browser and platform, and it should be mastered in the aspect ratio of the feed it is made for rather than cropped later, so a vertical 9:16 or square 1:1 version is framed as such at the shoot. Bit rate should be high enough that motion stays clean after the platform re-encodes it, which it always does, and captions should be supplied or burned in because most feed video is watched with the sound off. The podcast should be recorded and edited as uncompressed WAV at 44.1 kHz and 24 bit, then exported to MP3 at around 128 kbps for distribution, because podcast listeners are usually on mobile data and speech survives that compression well, while editing a lossy file would compound its artefacts. Running through all three is the same principle: the master is kept lossless and layered, and each delivery file is a separate export matched to its destination. Judging fitness for purpose means asking what the file has to survive, a printing press, a platform's own re-encoding, or a listener's data allowance, and choosing the format that survives it.