File Comparison Options
Duplicate File Detective provides a huge array of file comparison options. Find duplicates by name, size, file contents, music tags, and much more!
Nine independent comparison criteria are available - file names, file extensions, file sizes, file contents, last modified date, music tags, image metadata, document properties, and parent folders. Enable as many as you like: files must satisfy every criterion you switch on before they're grouped together as duplicates. Comparison options are stored with each project, so a photo-hunting project and a document-hunting project can each match files their own way.
Comparing files by content is the most rigorous approach, and you control exactly how it's done. Choose a content checksum algorithm to suit the job - CRC32 or ADLER32 for speed, MD5 for a 128-bit balance, or SHA-1, SHA-256, and SHA-512 for progressively stronger 160-, 256-, and 512-bit hashing. When absolute certainty is required, enable byte-for-byte content match confirmation to validate every candidate match at the binary level. Zip archives can even be compared by the files they contain rather than by the archive itself, with password support for encrypted archives.
Content hashing isn't always the right tool, though - and that's where Duplicate File Detective pulls ahead. A photo that has been recompressed, converted, or resized has completely different bytes, so its checksum will never match. The perceptual (visual) comparison basis solves this by comparing decoded image content instead, catching duplicate photos that look the same no matter how they've been re-encoded. Image metadata matching offers another route, grouping photos by date taken, camera manufacturer and model, and rotation-independent dimensions.
Duplicate songs present the same problem, and music tag comparison answers it: match by title, artist, album, track, year, genre, duration, and bit rate across MP3, FLAC, Ogg Vorbis, MP4/M4A, WMA, WAV, AIFF, APE, WavPack and more, regardless of how each file was ripped or encoded. Document properties do the same for your paperwork - title, subject, page, word and character counts, embedded creation date, and even shared PDF document lineage will match re-saved copies of a document whose checksum has drifted.
File name matching is every bit as flexible. Compare complete names, only the first or last N characters, or ignore trailing characters entirely. Match names whose words appear in any order, so "Beach Sunset.jpg" finds "Sunset Beach.jpg". Ignore whitespace, punctuation, or digits. And with the ignore copy and version decorations option, the artifacts Windows leaves behind - "- Copy", "(2)", "Copy of" - are stripped before comparison, so report.docx and report - Copy.docx land in the same duplicate group. Add parent folder matching to any depth from 1 to 99 folders to require that copies share the same folder context as well.