Compress Image

What Image Compression Can and Cannot Win

The only number that matters here is the one printed after the job finishes: source size, output size, and the change between them. Everything on this page is about making that number better, and about recognising the cases where it cannot be improved at all.

Where the savings come from

A JPEG or WebP encoder throws away high-frequency detail the eye is bad at noticing — the noise in a blue sky, the subtle variation inside a shadow — and spends its bits on edges and flat regions instead. Photographs contain enormous quantities of that discardable detail, which is why a 6 MB camera file often drops to 800 KB with no visible change.

Screenshots, line art and logos contain almost none. Their content is sharp boundaries and flat colour, exactly what lossy encoders handle worst, so lowering quality produces smeared halos around text long before it produces a small file.

Where to put the slider

Between 75 and 85 percent is the useful band for photographs. Below 60 the 8x8 blocks the encoder works in start showing, first around high-contrast edges. Above 90 the file grows quickly while the visible improvement disappears. Encoding runs in a background worker, so a 24-megapixel photo does not lock up the tab while it is processed.

PNG will disappoint you here

PNG compression is lossless. It looks for repeated pixel patterns and stores them once, and it never discards information, which is precisely why the quality slider does nothing to a PNG. A photograph saved as PNG has very little repetition to exploit, so it stays large.

Switch the output format to WebP and the same picture typically falls by 60 to 80 percent. Keep PNG only when you need transparency without loss, or when the sharp edges of a diagram or screenshot must be preserved exactly.

When the result comes back larger

This happens, and it is not a fault. If the source was already compressed hard — anything downloaded from a social platform, for instance — a second pass at a higher quality than the original spends more bits than it saves. Quality is stepped down and the encode retried; if the output is still bigger, you are told by how much and offered a button that hands back the untouched original.

Take the original. Re-encoding an already-lossy file layers artefacts on top of artefacts and buys nothing.

Metadata leaves with the pixels

Re-encoding rebuilds the file from decoded pixels, so camera make, lens, capture time and GPS coordinates do not survive the trip. That is usually welcome. If you wanted to read those fields first, or remove them without re-compressing anything, Image Metadata Viewer does both and leaves the pixel data byte-identical.

Order of operations

When the image is destined for a web page at 800 px wide, scale it down before compressing rather than after. Fewer pixels is a larger saving than any quality setting can deliver, and the encoder then works only on the pixels that will actually ship. Resize Image handles that step.

If a PDF is heavy because of the photographs inside it, shrink those images and rebuild the document, rather than rasterising every page through Compress PDF and losing the selectable text along the way.