Skip to content

8 formats documented

A reference for the formats document work runs through.

Choosing a format badly is the most common cause of a file that is ten times larger than it needs to be, or a photograph that looks like it was faxed. Each page here explains what a format actually is, then says where it stops being the right answer.

These are explanatory pages rather than tool pages. Where a conversion genuinely works in a browser it is linked; where it does not, the page says so instead of sending you to something that will disappoint you.

Side by side

Every format at a glance

Compression character is read from each format’s own specification table, so it cannot contradict the detail page. A dash means the question does not apply — PDF is a container, and what matters is how the images inside it were compressed.

The short answer

Match the format to the content, not to habit.

Nearly every format complaint traces back to one mismatch: a photograph stored in a lossless format that cannot compress it, or a screenshot of text stored in a lossy one that smears every letter edge. The formats are not competing — they were designed for different pictures.

Continuous tone
Photographs and scans of photographs want lossy compression. Their detail is noisy enough that discarding some of it is invisible.
Hard edges
Text, diagrams, screenshots and line art want lossless compression. Their detail is exactly the thing lossy encoders throw away first.

For a document that must look identical to everyone who opens it, the answer is PDF — and then the question becomes how the images inside it were encoded, which is where the pages above become useful.