Question 1

How does DuplicateDuster find duplicates?

Accepted Answer

By content, not by name. Files are hashed with SHA-256, and files with an identical hash are byte-for-byte identical — no ambiguity, no false match. Hashing runs across a pool of worker threads sized to your machine, so a folder with hundreds of thousands of files doesn't sit on one core.

Question 2

What are the "source" and "target" folders for?

Accepted Answer

They are what makes the tool safe to run on real data. Source folders hold the copies you're keeping — your photo library, the canonical archive. Target folders are the ones being cleaned. Matches are only ever paired across the two, and only files on the target side can be selected or removed. The source copy is shown next to every match, badged "keep", with no checkbox at all.

Question 3

Will it catch a renamed copy of the same file?

Accepted Answer

Yes. Because matching is on the file's content hash rather than its filename, a renamed or relocated copy is still recognised as a duplicate of the original.

Question 4

Can it find duplicate folders, not just files?

Accepted Answer

Yes. When every file under a target folder also exists in the source, the folder itself is reported as one finding instead of hundreds of separate file matches. Only the topmost match in a nested run is shown — if `photos/` matches `backup/photos/`, then `photos/2024/` necessarily matches too, and listing both would bury the useful result under its own children.

Question 5

Does it catch near-duplicates — a re-saved or slightly edited version of a file?

Accepted Answer

No, and we'd rather tell you that plainly than oversell it. DuplicateDuster matches exact content only. A re-encoded video or a re-saved document with even one changed byte will not be flagged. If you need fuzzy or similarity-based matching, that is not what this tool does today.

Question 6

How fast is it on a big drive?

Accepted Answer

Faster than it sounds, because most files are never read. Both sides are indexed by exact byte size first, and a file whose size occurs nowhere on the other side cannot be a duplicate of anything there — so it is skipped without being opened. Only the survivors get hashed, across a worker pool using about three quarters of your cores (capped at sixteen). On a mechanical drive, where parallel reads lose to seek time, the thread count can be pinned down to one.

Question 7

Is it safe — will it delete files I need?

Accepted Answer

Scanning is read-only and changes nothing on its own. Nothing is removed without an explicit confirmation that names the count and the total size. System locations — Windows, Program Files, /usr, /etc and their equivalents — are skipped during the scan and refused again at delete time whatever is selected, and the recycle bin is the default reclaim mode precisely because it is the reversible one.

Question 8

What are my options for reclaiming the space?

Accepted Answer

Four. Recycle bin, the default and the only reversible one. Permanent delete, for when the bin itself is the problem on a full drive. Secure erase, which overwrites before removing. And hard-link, which replaces the copy with a link to the original so both filenames keep working while the duplicate's blocks return to the filesystem — the useful option when something else depends on the path existing.

Question 9

What does "secure erase" actually do?

Accepted Answer

It overwrites the file's data in three passes (0xFF, then 0x00, then random bytes), renames it several times to scrub the original name from the directory entry, zeroes its timestamps, and then unlinks it. On Linux it hands off to `shred` where that is available. This is a meaningful step up from a normal delete, but it is not a certified military standard — and on an SSD, wear levelling and TRIM mean any overwrite-based erase offers a weaker guarantee than it would on a spinning disk. For flash storage, full-disk encryption is the stronger control.

Question 10

What does the free version include?

Accepted Answer

All of the scanning. Matching, folder detection and the reclaimable-space report are unlimited on the free tier, because the point is to prove the number is real. What the free tier caps is acting on it: up to 50 files per cleanup run. Pro removes that cap.

Question 11

Can I narrow what gets scanned?

Accepted Answer

Yes — minimum and maximum file size, an include list or an exclude list of extensions, skip hidden files, and skip system folders. Zero-byte files are always ignored: they match each other by definition and removing them frees nothing. The filter panel badges itself whenever a filter is narrowing the scan, so a setting left on from a previous run can't quietly explain a thin result.

Question 12

What happens to folders left empty after a cleanup?

Accepted Answer

They're removed, deepest first, so a nested tree collapses in one pass — but only if they are genuinely empty. A directory still holding anything you chose to keep is left exactly where it is.

Question 13

Does any data leave my machine?

Accepted Answer

No. Scanning, hashing and deletion all happen locally, and none of it needs a network connection. No file names, hashes or content are ever transmitted. The only outbound calls the app makes are validating a licence key and checking for a new version — and machines that never go online can activate from an offline licence file instead.

Question 14

What platforms does it run on?

Accepted Answer

Windows, macOS and Linux, from one codebase. The interface ships in English, Serbian (Cyrillic and Latin) and Russian. The 1.0 beta is not code-signed yet, so SmartScreen and Gatekeeper will warn on first launch.

Question 15

How is it different from Space Wizard?

Accepted Answer

Different jobs. Space Wizard answers "where did my disk go?" and clears duplicates as a side-effect of browsing. DuplicateDuster is a dedicated deduplication tool: a two-sided compare with SHA-256, whole-folder matching, and four ways to reclaim the space including hard-linking and secure erase. Use Space Wizard to find the problem; use DuplicateDuster when deduplication is the problem.

Questions we hear most.

Still have a question?

Talk to the team that actually builds the software.

Send us a message