Small files. Checkable results.

Case 01 · Image resolution

Can the same file print at 6 × 4 and 12 × 8 inches?

The sample PNG is a labeled 1800 × 1200 pixel grid. Its pixel count is fixed. Doubling both physical dimensions halves the effective resolution; changing a metadata DPI field would not create additional detail.

1. Download the PNG, then open the Image DPI Checker. You can also use its built-in sample button.
2. Select a 300 PPI target. Enter width 6 and height 4, in inches. Both axes should show 300 PPI.
3. Change width to 12 and height to 8. Both axes should show 150 PPI and the result should say it is below the 300 PPI target.
4. Try 6 × 6 inches. Width is 300 PPI but height is 200 PPI, so the tool should flag an aspect-ratio mismatch.

What should you change?

If 12 × 8 inches is required at 300 PPI, seek an original with at least 3600 × 2400 pixels. If only the current file exists, reduce the print dimensions or discuss a lower resolution with the supplier. For the square layout, decide whether to crop the image or retain borders; do not stretch it just to make the numbers fit.

The grid demonstrates dimensions, not photographic quality. A blurry or heavily compressed image can meet a pixel-count target and still print badly. This tool does not inspect focus, color profiles or embedded metadata DPI.

Reference: Adobe: image size, resolution and resampling. Our numerical examples are calculated from the sample dimensions.

Case 02 · PDF geometry

Why does one page look smaller even though its paper size matches?

A PDF can describe a full physical page using MediaBox and a smaller visible region using CropBox. Rotation is another independent value. The sample deliberately combines these differences so a single “all pages match” message would be inadequate.

1. Open the sample in the PDF Page Size Checker.
2. Expect three pages and two MediaBox size groups. Groups use dimensions rounded to 0.01 inch; they are not a print-tolerance certification.
3. Compare pages 2 and 3. Both have a Letter MediaBox, but page 3 has a 7.5 × 10 inch CropBox and 90° rotation.
4. Download the all-page report and compare these values before combining this PDF with another job.

What should you change?

Ask whether the job should use A4 or Letter and whether the crop is intentional. Re-export from the source layout where possible. Scaling a Letter page to A4 is not a neutral repair: margins, effective image resolution and text size may change. A rotation entry also does not mean the artwork itself is incorrectly oriented.

The checker reports physical MediaBox and CropBox dimensions, including PDF UserUnit scale. It does not inspect TrimBox, BleedBox, fonts, image resolution, color spaces or PDF/X conformance. Page geometry alone cannot prove that background artwork extends past the trim edge.

Reference: Adobe: cropping PDF pages. Cropping is not equivalent to rebuilding the source artwork.

Case 03 · Print planning

A 90 × 50 mm card with 3 mm bleed

This is a worked dimension example, not a universal printer specification. Suppose your supplier requests a 90 × 50 mm finished card, 3 mm bleed on every edge and 300 PPI raster artwork.

1. Open the Print Bleed Calculator and select millimetres.
2. Set trim width to 90, trim height to 50, bleed to 3 and resolution to 300.
3. Expect artwork size 96 × 56 mm: add 3 mm twice to each trim dimension.
4. Expect a raster canvas of 1134 × 662 pixels. The tool rounds each dimension upward to avoid falling below the requested size at the chosen PPI.

Bleed is outside; safety is inside

Extend background colors or images into the outer bleed region. Put names, logos and important text inside the trim, with the separate safety margin specified by your supplier. Do not assume a 3 mm bleed automatically defines a 3 mm inner safety margin.

Download the bleed plan to record the inputs. This calculator neither edits your file nor checks that artwork reaches the edge. For a vector PDF workflow, preserve the exact physical dimensions in the design document; the pixel canvas is only for raster artwork. Crop marks and their required offsets are separate supplier decisions.

Case 04 · EPUB conversion

Check content and reading order, not an assumed page count

Our DRM-free EPUB has a standard container file, a package manifest and two spine entries. Its original text was written for this test. It contains no illustrations, complex tables or protected content.

1. Choose the sample in the EPUB to PDF tool. Expect a ready message reporting two readable chapters.
2. Choose A4 or US Letter, then open the print view. Allow the popup if your browser blocks it.
3. Check that Chapter One comes before Chapter Two and the final sentence reads: “Page count depends on your browser and print settings.”
4. In the browser print dialog choose Save as PDF. If no dialog opens, use the print button in the new view or the browser’s Print command.
5. Open the saved PDF and confirm both chapters are readable. Do not overwrite your original EPUB.

What is not preserved?

This tool extracts a readable text layout rather than reproducing a publisher’s page design. Images, rich table structure, embedded fonts and original styling are not preserved by this EPUB path. It is unsuitable for illustrated textbooks or fixed-layout picture books. Page count varies with fonts, browser and print settings; we do not publish a fixed page-count claim.

If a real EPUB fails but this sample succeeds, the real file may be incomplete, encrypted or structurally different. Recovery from a missing container file is best-effort, not proof that all original chapters were recovered. Compare against an EPUB reader before using a recovered copy.

Case 05 · Recovery versus missing content

Missing metadata is not the same as a missing chapter

These variants come from the same two-chapter sample above. One removes only META-INF/container.xml; the other removes the second chapter file but leaves the reading order pointing to it. Comparing them isolates the structural problem without changing the story text.

1. Open EPUB to PDF. Choose either downloaded file, or expand its test-file menu to load the same cases directly.
2. For missing container metadata, expect two recovered text chapters and a visible review warning. Expand the chapter-order preview and verify Chapter One precedes Chapter Two.
3. In its print view, check that the final sentence still reads: “Page count depends on your browser and print settings.” Recovery has found the sample text, not repaired the original EPUB archive.
4. For the missing-chapter variant, expect an error identifying OEBPS/two.xhtml. The print action must stay disabled. A one-chapter output labelled successful would be an incomplete conversion.

What should you change?

Re-download or re-export an incomplete book. Do not delete reading-order entries merely to silence the error. If no package is available and the tool falls back to filename order, compare every section with the original reader; filenames cannot establish the intended reading order. An image-only section may have no extractable text and is reported as an omission. These tools do not perform OCR or remove DRM.

Reference: W3C EPUB 3.3: container, manifest and spine. These intentionally broken fixtures are recovery tests, not conforming EPUB examples.

Case 06 · Exact units

Why an A5 raster canvas needs 1819 × 2552 pixels here

Assume your printer wants A5 trim, 3 mm bleed on every edge and 300 PPI. In the bleed calculator, enter 148 × 210 mm, bleed 3 and PPI 300. Keep full precision until the final pixel result.

Rounding the physical dimensions to 6.06 × 8.50 inches first gives a smaller 1818 × 2550 pixel result. That is an intermediate-rounding error, not a different bleed requirement. The calculator rounds final pixels upward; vector layouts should retain the exact physical artwork size of 154 × 216 mm.

Now change only bleed to 3.175 mm, the exact equivalent of 0.125 inch. Expect 154.35 × 216.35 mm and 1824 × 2556 px. Do not substitute this larger canvas unless your supplier requests or accepts it. See the unit comparison and the folded-card setup example.

What this test bench verifies

Automated checks cover PPI and bleed arithmetic, invalid numeric inputs, the numerical tables in both bleed guides, PDF sample page geometry, report text and EPUB sample packaging. They also check server-rendered routes and canonical URLs. Browser regression scenarios exercise file selection, sample results, incomplete EPUB rejection and the printable view. These are bounded checks, not proof that every file or device is supported.

We have not certified commercial press output, color accuracy, accessibility of converted PDFs, or physical-device mobile printing. Saving through a native print dialog depends on the browser and operating system. The samples above let you repeat known cases on your device. If results differ, use Contact with the tool, browser version, non-sensitive inputs and observed result.

September 12, 2026 changes

• Corrected A5 pixel arithmetic and folded-card setup examples; added regression checks for article tables.
• Added controlled EPUB recovery and missing-chapter samples.
• Stopped conversion when a referenced EPUB chapter is absent; added recovery warnings and an extracted chapter-order preview.

September 10, 2026 changes

• Added local sample files and downloadable image, PDF geometry and bleed reports.
• Separated pixel-count results from image-quality claims and added aspect-ratio warnings.
• Separated PDF MediaBox, CropBox and rotation; documented checks that are not performed.
• Clarified browser-print conversion before file selection.