Turn on Duplicates and Kamembert searches every scanned drive together, not one at a time, so a file copied from C: to D: still shows up as a pair. It works in two passes:
- Quick pass: group files by name and size. This is fast and catches the obvious copies without reading file contents.
- Verify pass: before anything is deleted, the copies are read byte for byte to confirm they really are identical. Matching by name and size alone is never enough to remove a file.
Whole duplicate folders are detected too, and their files are not counted twice. Very small files are skipped by default, since build artifacts and config files come in large same-size swarms that add noise without freeing much space (you can change the size threshold in settings).
Near-duplicate folders
Kamembert can also flag folders that are nearly identical rather than exactly the same, which is common with copied projects. You set how similar they have to be: 50%, 70%, or 90% of the smaller folder. When a near-duplicate pair is reported, the exact duplicates sitting inside it are rolled up so the list stays readable.
Verified before delete. Removing a duplicate reads both copies in full to be sure they match, then moves the extra one to the Recycle Bin. Large folders are verified in the background so the window never freezes.