Like X-1a, but colour-managed RGB is permitted.
| ISO reference | ISO 15930-3:2002, revised as ISO 15930-6:2003 |
|---|---|
| Based on | PDF 1.3 for the 2002 revision, PDF 1.4 for the 2003 revision |
| Purpose | Print exchange |
| Transparency | Not allowed, must be flattened |
| Colour spaces | CMYK, spot, plus colour-managed RGB and Lab |
PDF/X-3 is the middle standard. It keeps most of PDF/X-1a’s discipline but makes one significant concession: colour-managed RGB and Lab content is permitted, as long as it carries an ICC profile so the conversion is defined rather than guessed.
Under PDF/X-1a, an RGB image is simply not allowed. Under PDF/X-3, it is allowed provided the colour is managed — the file must carry enough information for the RIP to convert it predictably, and an output intent must state the intended printing condition.
That matters most for photographic work. Converting to CMYK early throws away gamut you can never get back. PDF/X-3 lets the conversion happen later, at the RIP, using the printing condition that will actually be used.
Everything else from X-1a stays. Transparency must be flattened. No layers. Fonts must be embedded. No encryption, no external references.
That is the awkward part of X-3, and the reason it is less common than the other two: it modernised the colour rules but not the transparency rules, so you still take the destructive flattening step.
Most workflows have moved past it in both directions. Shops wanting maximum certainty stayed on PDF/X-1a; shops wanting a modern workflow moved to PDF/X-4, which allows live transparency as well as managed RGB. PDF/X-3 remains in use where a printer standardised on it years ago and has no reason to change.
If your printer has specified X-3, use X-3. If nobody has specified anything and they accept it, X-4 is the better default.
Conformance is not something you can confirm by eye. Upload a PDF and Filecheck reports which standard it actually conforms to, and what would have to change. Free, no signup, files deleted after processing.