EPUBTurnPDF to EPUB converter

Build a reproducible PDF to EPUB comparison record

By EPUBTurn · Updated

Record rights, input checksum, versions, settings, timing method and specific observed problems. This article provides a method, not unperformed benchmark results.

On this page
  1. Keep one published input stable
  2. Record the route and any intervention
  3. Define checkpoints before counting results
  4. Publish a result that another reader can inspect
  5. Use the record for a practical decision
  6. Source and licence
  7. Record an automatic sample as its own result
  8. NASA source

Keep one published input stable

Use the same authorised PDF for every route in a comparison. Our published research-paper comparison uses the same ten-page published research excerpt, with its exact file hash and byte size recorded. The source is a selectable-text scientific paper with two-column prose and vector plots, not a table-grid or scanned-book test.

Record the source title, edition, licence and extracted page range before conversion. Changing the input to suit one tool would make the results harder to interpret. Keep the original source separately from any preparation or excerpt.

Record the route and any intervention

Write down the tool version where available, selected mode and settings. Keep the downloaded output unchanged and record its hash. If you repair a crop, paragraph or navigation link, save that as a separate reviewed result and describe the intervention.

In our paper run, four competitors produced default downloads. EPUBTurn needed three manually corrected equation crops before a reviewed export was built; normal cloud download remained blocked. CloudConvert and Zamzar could not proceed with the new paper because existing free allowances were exhausted. Access gaps are distinct from inspected conversion defects.

Define checkpoints before counting results

Choose an exact column transition, a meaningful numerical association and a complete figure. For the paper, those are the page-two passage, the protected and fished population estimates, and the three panels of Figure 1. State the expected source relationship before inspecting any output.

Use retained, partial, issue or not available with a short observation. The reviewed EPUBTurn result keeps the checked passage and number associations, but the plot has clipped lower axis labels. A numeric match or retained caption cannot stand in for the complete relationship.

Published PDF

Source Figure 1: map, density plot and scatter plot with complete axes.
Source Figure 1: map, density plot and scatter plot with complete axes. Original PDF page. Open full-size image

Reviewed EPUBTurn result

Reviewed result retains the three panels but clips lower axis labels. Three equation crops were corrected before export.
Reviewed result retains the three panels but clips lower axis labels. Three equation crops were corrected before export. Browser view of the saved EPUB with its original styles; not a physical-reader test. Open full-size image

Publish a result that another reader can inspect

Provide the source, saved outputs, settings and screenshots from identifiable applications. State whether a screenshot is a browser rendering or a native reading app. Keep the original styles and record view settings rather than repairing one provider’s presentation for the comparison.

No speed ranking or population error rate follows from a single file. A valid EPUB package does not establish correct wording, complete figures or physical-device compatibility. Preserve untested areas in the report and avoid treating an unavailable output as a zero-quality score.

Use the record for a practical decision

Decide which failure would matter to the task you are trying to complete. A missed axis title can prevent chart interpretation even when prose is comfortable. A wrong column transition can change the meaning of an estimate without losing a digit.

EPUBTurn is a PDF reading-conversion route. Try a permitted sample, record which difficult features appear and compare them with the source. Check the full file separately if you continue; automatic sample selection does not verify every page.

Source and licence

Source: Hadj-Hammou et al. (2024), Global patterns and drivers of fish reproductive potential on coral reefs. CC BY 4.0. Pages 1–10 extracted; converted files and screenshots are adaptations.

Record an automatic sample as its own result

The actual NASA sample EPUB covers excerpt pages four to six, not every source page. Save it with the NASA report excerpt and record that automatic selection, lack of manual edits and remaining chapter/list formatting limits. The paper’s reviewed export is a separate record with three corrections. Keeping both scopes prevents a clean sample from being reported as a full-document benchmark.

NASA source

D. Harold Shelton, Apollo experience report: Lunar module stabilization and control system, NASA, 1975. US government work; public use permitted. Original pages 3–8 extracted; sample selects excerpt pages 4–6.

Files used in this guide

Download the source files, conversion results or worksheets referenced here. Each label identifies the source document or the converter used.

Published research-paper PDF (pages 1–10)EPUBTurn reviewed result; three equation crops correctedNASA technical report source excerptEPUBTurn actual NASA sample: three selected pages

Related guides