Guides

How to find duplicate files reliably

Two files can have the same name and different content, or different names and identical content. Here's how to find real duplicates reliably, by size and by fingerprint, instead of guessing from filenames.


You know there are duplicate files eating your space, but finding them by hand is hopeless. The instinct is to sort a folder by name and look for repeats. That's exactly where most duplicate hunts go wrong, because a filename tells you almost nothing about what's inside.

The filename lies in both directions

When you copy a file, your system renames it: photo (1).jpg, photo copy.jpg, photo-final-v2.jpg. The content is identical, but the names all differ, so sorting by name walks right past them. It happens the other way too. Two completely unrelated files can both be called scan.pdf or IMG_0042.jpg. Trust the name and you'll flag files that have nothing in common.

Size is a filter, not an answer

Comparing file sizes is a smart first step. If two files differ by a single byte, they aren't identical, so you can skip the comparison entirely. But a matching size proves nothing on its own. Plenty of different files happen to weigh the same, especially small ones. Size is there to shrink the pile quickly, so the real check only runs on candidates that could actually match.

The fingerprint settles it

The only reliable way to know two files are the same is to look at their content. A good tool reads the bytes and computes a short fingerprint from them. Same content gives the same fingerprint, every time, whatever the file is named. Different content gives a different fingerprint, so nothing is falsely paired. This is what separates a true copy from a file that merely looks similar, and it's worth knowing the difference: see exact duplicate versus similar file.

Doing it across a NAS

On a NAS the same logic applies, but the files sit behind a network share. NAS Ranger connects over standard SMB, filters by size first, then confirms by fingerprint, across whole shares at once. You get a clean list of real duplicate groups with the exact space each one would free, and you decide what goes.

Try it free: point it at your NAS, let it find every genuine duplicate by content, and see what you can reclaim before touching a thing.

FAQ

Why can't I just sort by name to find duplicates?
Because names lie in both directions. A copy is often renamed to photo (1).jpg or photo-final.jpg, so identical files get different names. And two completely different documents can share a name like scan.pdf. Sorting by name misses the first kind and falsely flags the second.
Is comparing file size enough?
Size is a good first filter, not a verdict. Many different files happen to share a byte count, and a single changed pixel gives two near-identical photos different sizes. Size narrows the field fast, then a content check confirms.
What does 'find by content' actually mean?
The tool reads the bytes of each file and computes a short fingerprint from them. Two files with the same fingerprint hold the same content, whatever they're called. It's the only method that's both certain and quick at scale.
Does this work for files on a NAS?
Yes. NAS Ranger connects to your NAS over standard SMB and applies the same size-then-fingerprint method to whole shares, so you find duplicates across the network, not just on your local disk.