Convert JPG to TIFF

It cannot improve the picture. JPEG works by discarding information the eye is least likely to miss, permanently, at the moment the file is saved. Converting to TIFF afterwards writes out every surviving pixel exactly — which is a real guarantee and a different one from the guarantee people hope for.

The distinction matters because of why TIFF is usually requested. If a journal or a printer asked for it in order to preserve image quality, converting a JPG satisfies the letter of the requirement and not its purpose: the quality they were trying to protect was already spent. If the original camera file or the export from the editor still exists, converting that is what they actually wanted. If it does not, this is the best available answer and is worth doing.

The value of TIFF is in what happens next. A photograph that goes through several rounds of editing, colour correction and resaving as JPG degrades a little each time — the artefacts compound, which is why images that have been through a few systems look the way they do.

Converting to TIFF stops that clock. Every subsequent save is lossless, so a file that passes through a designer, a proofing stage and a printer arrives as it started. That is precisely the reasoning behind the requirement, and it is a good one — the mistake is only in expecting it to work backwards as well as forwards.

TIFF is less a format than a filing system for images, and that flexibility is its weakness: it permits several compression schemes, and its readers agree on less than you would hope. LZW is broadly supported, Deflate is patchy, and JPEG-in-TIFF is a compressed image inside a TIFF wrapper, which is generally the opposite of what somebody asking for a TIFF wants.

The file written here is plain, uncompressed baseline TIFF — the version that every piece of professional software has read for three decades. The cost is size and the benefit is that it opens, first time, in whatever the recipient is using. For a format chosen specifically because it is dependable, that is the right trade.

Uncompressed means the arithmetic is simple: width times height times four bytes. A 12-megapixel photograph is about 48 MB whether the JPG it came from was two megabytes or six, because the JPG size depended on compression and the TIFF size does not.

That catches people out when emailing the result, and it is why submission portals that ask for TIFF usually also provide an upload mechanism built for large files. If the requirement is really about avoiding further loss rather than about the format itself, PNG is worth proposing: also lossless, readable everywhere, and typically half the size or less.

The format allows multiple pages, layers and 16 bits per channel, and none of that appears here, because none of it exists in a JPG to carry over. The alpha channel is the one exception, and it goes the other way: the writer declares four samples per pixel and flags the fourth as an extra sample, so the TIFF does carry an alpha channel and every value in it is fully opaque.

That channel is where the size arithmetic above comes from — four bytes per pixel rather than three. What comes out is otherwise a single-page, 8-bit image, which is exactly what a JPG is, written into a container that will not lose anything further. If the requirement involves layers or 16-bit depth, the JPG was never going to satisfy it and that conversation belongs upstream.

The TIFF is RGB, like the JPG it came from. TIFF can hold CMYK and print workflows often expect it, but a JPG is RGB, and inventing a CMYK separation here would mean guessing at an ink profile nobody specified.

If the recipient needs CMYK, that separation belongs in the tool holding the printer's profile. A correct RGB TIFF is a better starting point for them than a wrongly separated one.

TIFF carries tags of its own and can hold EXIF, but none of the JPG’s reach it. The picture is decoded to pixels and the TIFF is assembled from those pixels and a handful of structural tags the writer sets itself, so the date, the device and any coordinates stay behind in the JPG.

For a submission that is the right result and no further step is needed. For an archive it is the wrong one, because a TIFF kept as the record of a shoot arrives without the record. Keep the original JPG next to it if that history matters — nothing on this page can restore it afterwards, and this site’s metadata remover works on JPG, PNG and WebP rather than on TIFF.

Requirements that say TIFF often mean more than the extension: a resolution in pixels per inch, a colour space, sometimes uncompressed in particular. The extension on its own rarely tells the whole story.

Read the specification before converting rather than after being rejected. If it names a pixel dimension, a JPG smaller than that cannot be made to satisfy it by changing format — no conversion adds pixels that were never captured.

Drop a folder of up to a hundred and take the results as a ZIP. Nothing is uploaded, so there is no queue and no daily allowance.

The arithmetic is more of a constraint here than anywhere else on this site. Fifty 12-megapixel photographs are around 2.4 GB of TIFF being assembled inside your browser. Convert in small batches and save as you go.