Will the quality improve if I convert a photographic GIF to PNG?

The quality will not improve, but the current image is preserved without any further degradation. PNG is a lossless format; it stores the 256-colour image coming out of the GIF bit for bit. Banding and the dithering pattern remain visible, because that damage was done when the image was turned into a GIF. PNG's benefit is that no new compression loss occurs in any subsequent edit.

Which gives the smaller file?

On photographic content, JPG is smaller almost every time. JPG uses lossy compression to discard detail the eye struggles to notice; PNG discards no pixel information and only packs repeating patterns. That said, because the image coming from the GIF contains only 256 colours, PNG's compression efficiency can come out better than you would expect for an ordinary photograph. The dithering pattern still strains both formats, though, since an irregular dot structure is hard content to compress.

Which should I pick if there is a dithering pattern?

If the dithering — the scattered-dot pattern — is pronounced, PNG is the safer choice. JPG's compression logic cannot process that irregular, high-frequency structure efficiently; at a low quality setting the dots turn into blurry blotches and the image looks worse than the original. PNG preserves the pattern as it is. If you absolutely must use JPG, keep the quality setting high.

Does the decision change if the GIF has transparent areas?

Yes, and it becomes completely clear-cut: PNG. The JPG format has no concept of transparency; transparent regions are automatically filled with a flat colour and that cannot be undone. PNG both preserves the transparency and offers an alpha channel more capable than GIF's binary transparency. But since the GIF carries no soft-edge information to begin with, moving to PNG does not smooth the edges by itself — it simply preserves the transparency as it currently stands.

Converting a Photographic GIF: JPG or PNG?

The question is subtler than it looks

You have a GIF made from a photograph and you want to convert it into something more useful. The first reflex is usually "if it's a photo, use JPG," and in most cases that is right. But this particular situation is a little subtler, because the file you have is no longer an ordinary photograph — it is an image that has been through GIF's 256-colour limit and had its structure changed.

Deciding without understanding that difference can lead to needless file bloat or to further degrading the image. Below we compare the two on colour fidelity, file size, dithering behaviour and use case.

First, the shared truth: neither brings the loss back

There is one point to clarify before the comparison starts. The GIF format stores at most 256 colours per frame. That is a hard limit embedded in the format's file structure. The moment your photograph was turned into a GIF, an image of millions of tones was reduced to that palette: sky gradients became stepped bands, skin tones flattened, soft shadows terraced.

Neither JPG nor PNG brings that information back. JPG supports millions of colours, PNG is lossless — but neither can produce information that no longer exists. So your choice is not "which will look better" but "which will carry the current image better, and at what cost."

Colour fidelity: PNG wins outright

PNG is a lossless format. It stores the image coming out of the GIF bit for bit. Not a pixel changes, not a tone shifts. When you open the file you see exactly what was in the GIF's single frame.

JPG works lossily. When saving, it divides the image into blocks and discards detail considered hard for the eye to notice. On photographic content that is usually an acceptable trade. But an image that came from a GIF has a problem: dithering.

The dithering pattern strains JPG

During the conversion to GIF, the software applies dithering to hide the colour loss — scattering small dots to create the illusion of an intermediate tone between two colours. Up close it looks like a sandy texture.

That pattern is difficult content for JPG's compression logic. An irregular, high-frequency dot structure cannot be packed efficiently. One of two things happens: either the file comes out larger than you expected, or at an aggressive quality setting the dots run together into blurry blotches and the image looks worse than it did as a GIF.

PNG carries no such risk. It preserves the pattern exactly. On the other hand, because the pattern itself contains no repetition, it drags PNG's compression efficiency down too — so the file will not be small.

File size: JPG is clearly ahead

On photographic content, JPG produces a smaller file nearly every time. Lossy compression was designed for exactly this: packing content efficiently where every pixel differs from its neighbour and repeating patterns are scarce.

PNG discards no information and only compresses repetition losslessly. Because there is little repetition to pack in a photograph, the output is large. Since the image from the GIF has been reduced to 256 colours, PNG's efficiency can come out slightly better than for an ordinary photograph, but the dithering pattern eats that gain.

The practical conclusion: for an image you will publish on a website, send by email or keep many copies of, JPG's size advantage is real and meaningful.

Transparency: the one place PNG wins on its own

GIF supports binary transparency — a pixel is either fully transparent or fully opaque, with no value in between. In JPG the concept of transparency does not exist; transparent regions are filled with white or black during the conversion, and that cannot be undone.

If your GIF has transparent areas and you need to keep them, the comparison ends right there: PNG is the only option. But set your expectations correctly — moving to PNG does not smooth the edges by itself. Since the GIF carries no soft-edge information, the jagged edges remain exactly as they are. PNG only preserves the transparency that already exists.

Which one in which scenario

Photographic image to be published on a web page: JPG. The size gain is real and the visual difference is negligible. Keep the quality setting medium-to-high so the dithering does not blur.

A working file you will edit: PNG. If you do the cropping, text-adding and colour-correcting on a PNG, no new loss occurs at any save. You export the final version as JPG at the very end.

A frame containing text, interface elements or sharp edges: PNG. If a frame taken from a screen recording has writing in it, JPG will wreck it with halos and mess around the edges.

A GIF with transparent areas: PNG. There is no alternative.

An image with a pronounced dithering pattern: Prefer PNG. If JPG is unavoidable, save it at high quality.

Uploading to a system that only accepts JPG: JPG. This is compatibility, not preference. Here you can convert the GIF to JPG and finish the job in a single step.

Archiving, long-term storage: PNG. Storing losslessly makes sense so you do not lose the source, but the best move is to not delete the original GIF either — if there is animation, it lives only in that file.

In summary

Choose PNG if you need transparency or if the content contains text and sharp graphics. If you are publishing the image as-is and it is a photographic frame, JPG is both smaller and good enough. Neither brings back the colours GIF discarded; the choice is about the cost at which you carry the current image forward. If you are unsure, the simple rule is this: PNG if you are going to edit, JPG if you are going to publish.