重複ファイルの整理: what the free tier covers

Duplicate cleanup is one of the few Mac tasks where the software price is the least interesting number in the decision. The scanning itself is a solved problem that costs nothing: the operating system ships with everything needed to compare file contents, and the established open source utilities have no licence fee. What money buys is further along, in the part of the job that happens after the list appears. Anyone pricing 重複ファイルの整理 ryoukin is really deciding how much of the reviewing, selecting, and undoing they intend to do by hand, and that decision is worth making before opening a single product page.

What the Mac already includes

Nothing has to be bought to produce a list of duplicate files. A Mac arrives with the components already installed.

Finder can sort a folder by name, by size, and by date, and a smart folder can hold a saved search across a whole volume. For anything with an obvious naming pattern, such as files ending in a copy marker or a trailing number, that is often the entire job.

Terminal ships with utilities that compare file contents directly. Checksums can be generated for every file under a path and grouped, which is exactly what commercial scanners do underneath their interfaces. The result is a text list rather than a reviewable table, and building it requires assembling a few commands, but the detection quality is identical because the underlying comparison is identical.

Storage settings in System Settings handle a related problem at no extra cost: offloading files to iCloud, clearing caches, and emptying the Trash on a schedule. Those tools reclaim space without identifying duplicates at all, which is worth knowing before buying a duplicate finder to solve what is actually a disk usage problem.

Time Machine is included as well, and it is the reason any of this can be done safely. A backup that is current is what makes an aggressive deletion rule affordable, and it costs the price of an external drive rather than a licence.

Free of charge, in the open source sense

Beyond the built in tools sit the established open source duplicate finders, installed through a package manager. They carry no licence fee and no account, and several of them are considerably faster than a naive checksum pass because they prefilter by size and compare only the files that could possibly match.

Free of charge is not the same as free of cost. What these tools do not include is the part that consumes the time.

There is no review table showing both paths side by side with previews, so deciding which copy survives means reading paths in a terminal. There is no undo, which means the safety net has to be constructed deliberately, by writing results to a file first and acting on them second. There is no similarity matching for photographs, since that is a different class of problem from byte comparison. And the selection rules, if the tool has them, are expressed as arguments rather than as settings, which is either an advantage or an obstacle depending on how the rest of the work gets done.

For a one time cleanup of a few thousand files, this route costs an evening and nothing else. For a library of hundreds of thousands of photographs, the missing review interface is the entire difficulty, and no amount of command line skill removes it.

What a paid product is actually selling

The feature lists blur together, but the things that genuinely cost money to build fall into four groups.

Review interfaces come first. Showing two candidates with previews, letting a selection be adjusted before anything is deleted, and keeping that state across a session is a large amount of interface work that open source utilities generally do not attempt.

Selection rules come second. Automatically marking the copy in Downloads rather than in Documents, or keeping the shallowest path, is what converts an unreviewable list into an actionable one. A tool without them leaves every row to a human.

Similarity matching for images is third, and it is the one capability with no free equivalent that is easy to set up. Comparing photographs that were exported at different sizes requires a different algorithm from comparing bytes.

Reversibility is fourth: moving to the Trash rather than deleting, writing a log of what was removed and from where, and being able to reverse a batch that was run with the wrong rule. This is the least visible item on any feature list and the one most likely to matter on the day it is needed.

The three shapes a price comes in

Pricing pages in this category take one of three forms, and the difference shows up after the purchase rather than at it.

Shape	What happens after payment stops	Where the risk sits
One time licence	The installed version keeps working	Major versions may be charged again, and support for new macOS releases may require the upgrade
Subscription	The application stops working or reverts to a limited mode	Recurring cost continues whether or not the tool is used
Bundle subscription	Access ends for every application in the bundle	One cancellation removes tools that are unrelated to each other

For a task that is genuinely occasional, a one time licence is the shape that matches the usage. Cleaning duplicates twice a year and paying every month is the mismatch that produces most of the regret in this category.

Subscriptions make more sense when the tool is part of daily file work rather than an annual event, because the recurring payment is buying continuous updates against an operating system that changes every year. The question to answer honestly is which of those two descriptions fits, and the answer comes from how the duplicates are being created. If the source is constant, the tool gets used constantly.

Distribution channel changes the shape too. An application sold through the Mac App Store inherits that store's rules for refunds, family sharing, and renewals, while a direct download from the vendor is governed by whatever the vendor publishes. Refund windows in particular differ, and for a tool that has to be tested against a real library before its value is known, the refund policy is part of the price.

Licence counting is the other detail worth reading closely. Per machine and per person are different, and anyone running a laptop plus a desktop will pay twice under the first model. The terms are usually stated plainly on the pricing page once you know to look for them.

Where the free tier stops

Almost every commercial duplicate finder offers something at no cost, and the boundary is not in the same place across products. Read for the boundary rather than for the word free.

The common line is between finding and acting. Scanning is offered without payment, the results are shown in full, and deletion is what the licence unlocks. That arrangement is reasonable, because scanning is the cheap half, but it means a free tier cannot complete the job by itself.

Other products limit by volume instead, allowing a number of files to be removed before payment, or restricting which locations can be scanned. A third pattern is a time limited trial with no functional restriction at all, which is the most useful shape for evaluation, because it lets the tool be tested against the actual library rather than against a sample.

This is where a free scan still has value even if nothing is bought. Running it answers the question that determines the whole decision: how many candidates exist, and how many of them a location rule would resolve without a human reading the row. A few hundred rows is an afternoon of manual work. A hundred thousand rows is not, and that is the point where paying for selection rules stops being optional.

Operational limits sit alongside the price and are easy to miss. Whether external volumes and network shares are supported, whether Full Disk Access is required, and what the tool does with its log are the kind of details that live in a vendor FAQ rather than on the pricing page, and they decide whether the licence is usable for the library you actually have.

Storage is the competing purchase

The alternative to spending money on cleanup software is spending it on space, and for the disk usage case it is frequently the better trade.

An external drive is a one time cost that removes the pressure entirely, and it doubles as the backup destination that makes deletion safe. Additional cloud storage is a recurring cost that does the same thing for files that need to be reachable from other devices. Neither reduces the number of copies, which means neither helps if the real problem is search results rather than free space.

The way to choose is to name the symptom. Running out of room is a storage problem, and storage is the direct purchase. Not knowing which of five identical filenames is current is an organisation problem, and no amount of storage improves it. The second case is the one where software earns its price, and it is also the case where the cheapest fix might be free: deciding that one folder is authoritative and everything outside it is disposable.

Four things to confirm before paying

Price is not the only fact on a vendor page that changes. Mac utilities have a long history of being sold, retired, and rebuilt, and all four of these are checkable in a few minutes.

Confirm the product is still being sold rather than merely still downloadable. Confirm that new customers are being accepted, since some tools stay available to existing licence holders while closed to everyone else. Confirm whether the company behind it has changed ownership, which is common in this category and typically changes both the update rhythm and the pricing model. Confirm whether the price was revised recently, because a recent revision signals which direction the vendor is moving in, usually from one time purchases toward subscriptions.

A comparison of file managers is useful at this point for a different reason than feature matching: it shows which products are still actively maintained. Something that has not been updated across recent macOS releases is a risk regardless of how reasonable its price looks.

What to change first

Run a free scan against one folder and count the candidates before looking at any price. The count, and the share of it that a simple location rule would resolve, is the only input that tells you whether this is an afternoon of manual work or a purchase.

If the answer is a purchase, choose the pricing shape that matches how often the job recurs rather than the lowest headline figure, and check reversibility before features. Tools built around recurring file work rather than one time cleanups, such as Atriens, are priced on that assumption, so the match between usage and shape is the thing to get right.