Contents
Duplicate finder (identical files)
The duplicate finder tracks down files that are identical in content — equal byte for byte, no matter what they are called or where they sit. That frees the space wasted by copies stored more than once. It is part of the file organiser (chapter 9): you open the organiser, scan, and then search for duplicates within a category.
8.1 Searching for duplicates
- First analyse a job — only then can the organiser be opened.
- In the header bar, click the “Toggle file organiser” switch; the panel opens in the tile overview.
- Click 🔍 Scan — the files are sorted into categories.
- Open a category tile (🖼 Images, say) and click 🔍 Duplicates (tooltip “Search for duplicates in this category”).
The ◄ Both ► / ◄ Left / Right ► switch determines whether both folders are compared with each other or only one side within itself. Files that are exactly equal at the same relative path on both sides (a backup mirror) count as duplicates too — deliberately so.
8.2 How identical files are detected
So that even large collections are checked quickly, the finder works in three tiers — from cheap to expensive, so that the costly check only hits the few real candidates:
| Tier | Method |
|---|---|
| 1 · Size | Files of different size can never be identical — they drop out immediately |
| 2 · Partial checksum | Of the remaining candidates, the first and last 4 KB are compared as SHA-256 (quick pre-filtering) |
| 3 · Full checksum | Only the final suspects are verified completely by SHA-256 — that is the proof of “byte-identical” |
During the run you see the progress in the status line, for example “⚡ {n} size candidates”, “🔎 {n} partial checksums” and “🔍 SHA-256… {i}/{n}”. Very large files (over 2 GB) are skipped.
8.3 The result: groups and “keep”
Every find appears as a group of equal files: header “📋 Duplicate #N — X files · size · wasted: Y”, below it the individual copies with preview, folder path, size, date and side marker (◄ Left / Right ►). Above the list, a banner sums it all up (groups · files · wasted space).
In each group one file is kept (green, “keep”), the rest are earmarked for removal. By default the newest file is preselected. The header buttons change that per group in an instant:
| Button | Keeps… |
|---|---|
| ⏱ Newest | the most recently modified file |
| ◄ Left | the file on the left side |
| Right ► | the file on the right side |
| ✓ All | all of them — the group is not cleaned up |
If a group drops below two files after removal, it disappears automatically — a single file is no longer a duplicate.
Looking closely: the exclusive search
With many groups, clicking through becomes tedious as soon as your criterion is not “newest” or “left side” but hangs on the name — “I only want to see the groups containing ‘original’”, say. That is what the ⚡ Exclusive search button in the header of the duplicate view (next to the 📊 statistics) is for. It searches the files of all groups; how to use the window is described in chapter 5.5.
With ✓ Apply hits, only the groups containing at least one hit remain — and they remain complete, with all their copies. That is deliberate: a duplicate group without its siblings says nothing any more; you would see one file without seeing what it stands against.
8.4 Cleaning up, undoing, making it final
🗑 Clean up (tooltip “Remove unticked files (restorable with undo)”) removes the copies marked for deletion. What happens then depends on whether secure deletion is active in the bar (chapter 13):
- Without secure deletion: the files are not deleted straight away but moved into an internal recycle cache of the program. ↩ Undo (tooltip “Restore deleted files”) can therefore bring everything back.
- With secure deletion (1×/3×/7×): the files are overwritten several times in place — there is no undo here, because overwritten content is gone for good.
Before the scan the view reads “Click Scan to find duplicates”, during the run “Searching for duplicates…”. If the finder finds nothing, it reports “No duplicates found” and “All files in this folder are unique”.