Convert PNG to JPG

Why a PNG is so large

PNG is lossless. It stores every pixel exactly as it was and reconstructs the image perfectly. That is the right behaviour for a logo, a diagram, a chart or a screenshot of text, where a single blurred edge is a visible defect.

It is the wrong behaviour for a photograph. A photograph is full of subtle gradients and sensor noise that no viewer could describe from memory, and PNG faithfully stores all of it. That is why a photograph saved as PNG can be ten times the size of the same photograph saved as JPEG, with no visible benefit whatsoever.

This catches people most often with screenshots. Most operating systems capture the screen as PNG, so a screenshot of a photograph, a video still or a map inherits PNG’s size without any of its advantages.

What you lose, and when it matters

JPEG is lossy: it discards information human vision is poor at noticing. On a photograph, at a sensible quality setting, the loss is invisible and the saving is enormous. On flat colour and sharp edges it is the opposite — JPEG produces faint halos around hard boundaries and a mottled texture in areas of solid colour, both of which are most obvious on text.

So the rule is simple. If the image is a photograph, convert it and do not think about it again. If it is a screenshot of an interface, a logo, a diagram or anything containing text that needs to stay crisp, keep the PNG — and if it needs to be smaller, reduce its dimensions rather than changing its format.

Transparency will be flattened

JPEG has no alpha channel. It cannot store transparency at all, so anything transparent has to become a solid colour. FileBelow composites onto white by default and tells you that is going to happen before you process, not after. You can change the colour in Advanced options if white is wrong for wherever the image is going.

If you need transparency preserved and a smaller file, WebP is the answer: it supports an alpha channel and compresses far better than PNG.