PDF/X, Output Intent & Page Boxes — The Print-Ready Target

By now you understand color, ink, and presses. This section is about the file you actually hand to the printer. A file that looks perfect on your screen can still print wrong — fonts go missing, colors shift, shadows turn into white boxes, edges show white slivers. The solution is a special, locked-down kind of PDF built specifically for printing: PDF/X. We'll learn what it is, why transparency is the great dividing line between its flavors, what the "output intent" is, and how the five page boxes tell the press exactly where to cut.

9.1 Why PDF/X Exists — The "Blind Exchange" Idea

A normal PDF is built for viewing on a screen. A print PDF must be something stricter: a blind, deterministic exchange file. "Blind exchange" means the printer can open it and run it with no questions asked — no fonts to chase down, no color to guess at, no risky features to trip over.

PDF/X

Stands for "PDF for eXchange." It is a restricted subset of the normal PDF format, standardized by ISO (the umbrella standard is ISO 15930). Each "flavor" — X-1a, X-3, X-4 — is a separate part of that standard.

Restriction

PDF/X forbids things that make printing unpredictable (missing fonts, unmanaged RGB, JavaScript, encryption, audio/video) and requires things the printer needs (all fonts embedded, a declared output intent).

9.2 The Transparency Model & Flattening — The Great Divide

The single biggest difference between the PDF/X flavors is how they handle transparency. Transparency means any artwork that lets what's underneath show through: drop shadows, glows, soft edges, blend modes, and reduced-opacity objects.

Live (native) transparency

The PDF stores the actual transparency instructions — the blend modes, opacity values, drop shadows, and soft masks. The RIP (the press's processor) resolves them at output time, at the device's full resolution. This was introduced with PDF 1.4 (Acrobat 5 / InDesign 2, around 2001).

Flattening

A pre-processing step that destroys live transparency. It carves overlapping transparent artwork into a mosaic of fully opaque pieces — some kept as vector, some turned into raster images — that look like the original blend but contain no transparency anymore.

Why flattening was ever needed

The original PostScript imaging model — and the older RIPs built on it — has no concept of transparency at all. Every object is opaque. A RIP that only understands PostScript Level 3 cannot interpret a blend mode, so any transparency must be "baked in" before it reaches that older equipment. PDF/X-1a and X-3 require flattened files precisely so they run on this older, opaque-only generation of machines.

How the flattener works

Atomic regions

The discrete tiles the flattener slices artwork into wherever transparent objects overlap. Each tile is reduced to a single, fully opaque appearance (a vector fill or a raster patch).

Rasters/Vectors balance

A slider plus a flattener resolution setting controls how much becomes pixels vs. vectors. Typical values: line art & text at 1200 ppi, gradients & meshes at 300 ppi.

Clip Complex Regions

An option that forces region boundaries to land on object edges, reducing visible seams.

Flattening artifacts — the failure modes

• Stitching / hairline artifacts — faint white or black lines tracing the boundaries between atomic regions, caused by anti-aliasing where a rasterized tile meets a vector tile. Often visible on screen and sometimes in print.
• Text rasterized or outlined — small type sitting over a shadow can get rasterized at flattener resolution, turning slightly fuzzy or heavy.
• Color shifts at overlaps — spot colors near transparency can be converted to process builds, or RGB→CMYK can shift inconsistently across a region.
• Broken overprint / knockout — flattening changes overprint relationships, so objects print over or under each other incorrectly.

The modern fix: PDF/X-4 keeps transparency LIVE — no flattening at all. A modern Adobe PDF Print Engine (APPE) RIP resolves transparency natively at device resolution. The result is smaller files, no stitching seams, sharper text, more predictable color, and "late binding" of color. This is the reason X-4 is the modern default.

9.3 The Three Standards — Differences and When to Use Each

In plain language:

• X-1a — "Everything is already CMYK and flat. No color management, no surprises." The 20-year-old safe-but-dumb standard: total predictability, zero flexibility. Use it only when a printer explicitly demands it or for a legacy RIP.
• X-3 — "I can send RGB/Lab images and let an ICC profile convert them, but transparency is still flattened." Historically common in Europe; a transitional standard.
• X-4 — "Live transparency plus ICC color management." The modern recommended default for virtually all commercial and digital print. Required whenever the design uses real transparency (shadows, glows, blend modes) and you want it resolved at the RIP.

(There's also PDF/X-5 = X-4 plus external referenced profiles/graphics — niche — and PDF/X-6, the newest part based on PDF 2.0.) Industry consensus from the PDF Association: stop using X-1a; default to X-4.

9.4 Output Intent — The Embedded "Target Press Condition"

The output intent is a required PDF/X entry that declares the exact printing condition the file was prepared for — in plain words, "this file is built to look right on this paper / ink / press combination."

What it contains:

Subtype GTS_PDFX

The only output-intent subtype valid for PDF/X.

OutputConditionIdentifier

The reference name of a standard print condition from the ICC Characterization Data registry — e.g., FOGRA39, FOGRA51, CGATS21-CRPC6 (GRACoL), CGATS TR 001 SWOP.

DestOutputProfile

The embedded ICC profile describing that condition — e.g., Coated FOGRA39 (ISO 12647-2:2004), GRACoL2013 CRPC6, US Web Coated (SWOP) v2.

OutputCondition / RegistryName / Info

Human-readable text describing the condition.

The conditions it points at (with their standard total-ink-limit figures):

• FOGRA39 — coated stock, older European ISO 12647-2:2004; total area coverage (TAC) ~330%.
• FOGRA51 / FOGRA52 — newer ISO 12647-2:2013 (PSO Coated v3 / uncoated v3).
• GRACoL 2013 (CRPC6) — US premium coated; TAC ~310%.
• SWOP (CRPC5 / CRPC3) — US web/publication; TAC ~300%.

Why it matters: the output intent is the file's color contract. The RIP uses it as the destination profile for any color conversion and as the assumed proofing target. Without it, the file is not valid PDF/X. Note that PDF/X allows only printer (output-device) profiles here — unlike PDF/A, it never permits monitor profiles.

9.5 Page Geometry Boxes — Where the Blade Cuts

Every PDF page defines a set of nested rectangles called page boxes. The PDF spec requires only the MediaBox; the other four are optional in general PDF — but a real print PDF needs the TrimBox and BleedBox.

MediaBox

The only mandatory box: the full physical sheet/canvas. In prepress it's deliberately oversized to hold bleed, crop marks, and slug/registration info. Everything else must fit inside it.

CropBox

The region a viewer displays/prints; defaults to the MediaBox if absent. Avoid it in prepress — it can hide bleed and confuse imposition.

BleedBox

The clip boundary for production output; extends beyond the TrimBox so ink runs past the cut line. Must be ≥ TrimBox.

TrimBox

The finished page size after cutting (the A4 / Letter / business-card size in the customer's hand). The single most important box for print — imposition and finishing equipment read it to know exactly where to cut.

ArtBox

The meaningful-content extent; today often used to mark the safe / live area — keep text and logos inside it.

The containment rules: MediaBox ⊇ BleedBox ⊇ TrimBox ⊇ ArtBox. All of CropBox, BleedBox, TrimBox, and ArtBox must fit inside the MediaBox.

Bleed is not Trim. Trim is where the blade actually cuts; bleed is the extra image area pushed past the cut so a slightly-off blade never exposes white paper.

Canonical numbers

• Standard bleed = 3 mm (Europe / ISO) up to 0.125 in (⅛ in ≈ 3.175 mm) (US); typical range 3–5 mm ≈ 9–14 pt. Large-format jobs may use 5 mm or more.
• Safety margin (keep critical content inside trim): typically 3–5 mm / ⅛ in in from the trim line.
• PDF coordinates are in points (1 pt = 1/72 in); so 3 mm ≈ 8.5 pt and ⅛ in = 9 pt.

What each standard requires

• PDF/X-1a & X-3: require MediaBox + TrimBox + BleedBox (TrimBox mandatory; BleedBox present when there is bleed).
• PDF/X-4: requires MediaBox + (TrimBox OR ArtBox) — and a page must not have both TrimBox and ArtBox. TrimBox is the normal choice.

┌─────────────────────────────────────────────────────────┐  <- MediaBox  (full sheet; holds marks; ONLY required box)
│   +  crop marks / registration / slug live out here      │
│   ┌─────────────────────────────────────────────────┐   │  <- BleedBox  (ink runs to here; >= TrimBox; ~ +3 mm)
│   │::::::::::::::: bleed zone (image overruns) :::::::│   │
│   │::┌───────────────────────────────────────────┐::│   │  <- TrimBox   (* FINAL CUT line = finished page size)
│   │::│                                           │::│   │
│   │::│   ┌───────────────────────────────────┐   │::│   │  <- ArtBox    (safe / live area; keep text+logos in)
│   │::│   │   TEXT . LOGOS . CRITICAL CONTENT  │   │::│   │     (~3-5 mm in from trim)
│   │::│   └───────────────────────────────────┘   │::│   │
│   │::└───────────────────────────────────────────┘::│   │
│   │:::::::::::::::::::::::::::::::::::::::::::::::::::│   │
│   └─────────────────────────────────────────────────┘   │
│   +                                                      │
└─────────────────────────────────────────────────────────┘
   Nesting:  MediaBox >= BleedBox >= TrimBox >= ArtBox
   (CropBox = viewer display area; omit it in prepress)

9.6 Putting It Together — The Shop-Ready Checklist