PDF to PNG

How to use it

1. Choose a PDF — it stays on your device.
2. Keep PNG selected and pick a resolution (150 DPI is a good default; 300 for print-quality).
3. Convert. Each page becomes a PNG; download one, or all of them together as a zip.

Lossless deflate, and where it pays off

PNG keeps every pixel exactly, using per-row prediction filters followed by DEFLATE. Nothing is approximated, so the file matches what was rendered, pixel for pixel. On a page of black text on white that also happens to be efficient: the filters reduce long stretches of unchanging background to almost nothing, and an A4 text page at 150 DPI often lands in the low hundreds of kilobytes.

The same mechanism works against you on photographs. Every pixel of a photograph differs slightly from its neighbour, the filters find little that repeats, and nothing may be discarded to make up the difference - so a photographic page can come out ten times larger than a lossy encoding that no one could tell apart from it. PNG rewards pages made of text, tables, diagrams, screenshots and line art, and punishes pages made of pictures.

Choose PNG when something else will process the image

An exported page is often not the final artefact. It gets fed to OCR, dropped into a slide deck, annotated, cropped and saved again. Each save in a lossy format compounds the approximations of the one before it, and OCR suffers first: a recogniser reads glyph boundaries, exactly the detail a block-based codec blurs.

PNG also carries an alpha channel, which matters when part of a page is going to be lifted out and placed over a different background - a logo, a diagram, a signature block. Lossy formats have no transparency and flatten it onto white. And because the output is exact and deterministic, two PNG renders of the same page can be compared directly, which is a practical way to confirm that a revised document really does render the same as the one it replaced.

Questions