The new version PDiff 3.7.4 has been released. New features in this release of the text-based PDF comparison solution are among others:

  • NEW: In the case of PDF files, which incorrectly contained repeating letter groups in the extracted text, an improved result can be achieved with the new option combine duplicate segments under Settings > PDF > Text Extraction Flags. It is recommended to switch on the new option together with the option preserve character quads.
  • NEW (Add-On Images):
    • New Tab Settings > Images with the option Compare images and adjustable tolerance for the image comparison (Threshold thr and Blur filter σ)
    • New in the dialog window Compare images (after double-click on a table row with images of type RPL or EQ):
      • Interactive setting of the tolerance parameters for the image comparison (Threshold thr and Blur filter σ)
      • The impact of the current tolerance values ​​on the difference image is shown immediately and illustrated with a histogram
      • Immediate display of the current final result EQ or RPL, resp.
  • Update: Latest Adobe PDF Library version 18.0.5PlusP1c
  • Fix: CSV export at the first start of PDiff after installation did not contain any columns
  • Fix: Report Annotated PDF A+B could produce "visually" empty pages for certain input PDFs
  • Fix: Render flag underline recognition yielded an error message on loading certain input PDFs
Version 3.7.3:
  • NEW: For text with position changes of the type POS (see Settings > Comparison > Text coordinates must match), the table column Distance shows the translation dx, dy of the maximally des maximal moved word within the text block
  • NEW (Add-On Images):
    • Images can be clicked like words with a synchronized text cursor (PDF preview A and B in the upper half of the window, text column A and B in the lower half of the window)
    • Dialog window for displaying pixel deviations (Delta) for images recognized as replacements (type RPL): Double-click on the table line
    • Handling of images which were embedded in the PDF as inline images
    • Fix: Processing of embedded binary pictures (b/w pictures)
Version 3.7.2:
  • NEW: Opening of PDiff projects using a URL and diff id (e.g. pdiff:/path/to/project.pdiff#42)
  • NEW: Add-On Images (Beta) to highlight differences for:
    • inserted or deleted pixel images
    • optionally changed image positions, with adjustable tolerance (see Settings > Comparison > Text coordinates must match), translation dx, dy is displayed in a new table column distance

PDiff shows you quickly and reliably all text differences between two PDF documents – as a comparison of the two PDFs and with markings of the deviations. So you can check intended revisions quickly and find unintended changes at first glance.

PDiff easily outperforms document comparison with the naked eye in terms of speed and precision: a human reader achieves an average reading performance of 200 words per minute with an error detection rate of 75%. On the other hand, PDiff manages about 400000 words per minute – that’s 2000 times faster – and above all it works tirelessly with 100% accuracy.

Applications of PDiff are among others: legal documents and contracts, pharmaceutical packaging and medical leaflets, important business documents, technical manuals, financial reports, manuscripts, book texts as well as price lists and catalogs.

PDiff is available as 64-bit version for Windows and macOS. Both versions are based on the proven Adobe® PDF Library™. Business customers can request a free trial at https://www.csci.de/en/pdiff/trial.