How can I tell if a USB drive or memory card is fake before I pay for a whole batch?
Check the price against the memory content first, then test one unit destructively. A drive priced below roughly the cost of the NAND inside it is not a bargain — flash has a floor cost, and a product under that floor is reporting capacity it does not have. The physical test is a full write-and-read-back across the entire nominal capacity: fill it completely, read every byte back, and compare. Free tools like H2testw on Windows or f3 on macOS and Linux do exactly this. A genuine drive passes; a fake fails at the point where its real storage ends.
The price test comes before any testing
Before you spend hours testing, spend ten minutes on arithmetic. Flash memory has a real floor cost per gigabyte, and every drive contains memory plus a controller plus assembly. A price far below the sum of those parts is not a discount, it is a different product wearing the label.
This is not a rule about cheapness in general — a high-volume buyer can legitimately pay less than a retail shopper. It is a rule about the floor: below a certain point, the only way to reach the price is to put less memory inside.
The destructive test that settles it
Capacity fraud cannot be detected by looking at the drive. It has to be written past. The test therefore fills the drive completely and reads it back.
- Pick one unit per batch — the test takes hours on large capacities, so sample rather than test everything.
- Run H2testw (Windows) or f3 (macOS/Linux) across the entire drive. Both are free.
- Do not interrupt it. The test writes until the drive is full and then verifies every sector.
- Read the failure point. If it fails at a fraction of the nominal capacity, that fraction is roughly what is really inside.
- Cross-check: a genuine 32GB drive shows about 29–30GB usable, because capacity is quoted in decimal units and the file system uses binary, plus its own overhead. A drive showing dramatically less than expected is suspect.
What a fake actually does when you use it
A counterfeit controller is programmed to report a large capacity, and it wraps writes around when it reaches the end of the real memory. The first files written stay; later files overwrite them. Every write reports success.
That is why the failure is so damaging: nothing errors at the moment it happens. The user discovers it later, when reopening a file written early on, and by then the data in the middle of the drive may be gone. For anything involving patient records, client footage or irreplaceable photos, this is the single failure mode worth designing against.
Sources
Figures quoted on this page are from the published reporting listed above. Where a number is an estimate rather than a published figure, it is described as approximate.
Frequently asked questions
Can antivirus or a disk utility tell me if a drive is fake?
No. A counterfeit drive is not malware, and it presents itself correctly to the operating system — the size it reports is written in firmware, and the OS believes it. Disk utilities read the same reported capacity and will confirm the drive looks fine. Only writing past the real memory and reading it back exposes the fraud.
Is a fake drive always cheap?
No, and that is the trap. Counterfeits are often priced at a plausible discount rather than an obvious one, because an implausibly low price deters buyers. The price test narrows the field, but it does not clear anything — a drive at a normal price still needs the write-and-read-back test before you commit to a batch.
What should I require from a supplier to avoid this?
A written statement of the test method used on the batch — specifically full-capacity write-and-read-back, not a capacity read — plus a result per carton with a lot identifier. And a clause that any unit failing that test is replaced. A supplier who tests this way will describe the method without prompting; one who describes "full inspection" in general terms has probably not tested capacity at all.
Working on a specification?
Send us the product type, capacity, quantity and target date. We reply to business enquiries within 1 business day.
Request a quote