For brand owners and solution providers sourcing USB flash drives, TF cards, SSDs and portable SSDs (PSSD), the most common after-sales question is simple: “the card says 64GB, so why does my device show 59?” The label is not wrong and the product is not short of memory. Two independent effects stack up: a difference in how capacity is counted, and a small amount of space the file system reserves for itself. Understanding both lets a buyer write capacity class and acceptance into a purchase order so a normal display difference is never treated as a defect.
Two ways to count the same bytes
Storage capacity is quoted and marked in decimal units, where 1 GB equals 1,000,000,000 bytes (109). This convention sets the capacity classes on cards, drives and SSDs, and it is how the industry labels a “64GB” product.
Most operating systems, however, report space in binary units, where 1 GiB equals 1,073,741,824 bytes (230) — yet many still display “GB.” Because one binary gibibyte is roughly 7.4% larger than one decimal gigabyte, the same device produces a smaller number in binary.
Run the arithmetic on a 64GB class product: 64,000,000,000 ÷ 1,073,741,824 ≈ 59.6. A device therefore shows roughly 59–60 “GB.” That single calculation explains the large majority of industry-common complaints about “missing” capacity. The way to avoid it is to define the capacity class in decimal bytes, and to state the expected binary display in the purchase order.
| Capacity class (as marked) | Decimal bytes | Typical binary display after formatting | Why the display differs | How to write it in the purchase order |
|---|---|---|---|---|
| 64GB | 64,000,000,000 | ≈ 59.6 GiB (often shown as “59–60 GB”) | Decimal marking vs binary reporting, plus file-system overhead | Define the class as 64,000,000,000 bytes; pre-accept a binary display of about 59–60 GiB |
| 128GB | 128,000,000,000 | ≈ 119.2 GiB | Same two-step reduction | State the decimal byte value as the acceptance floor; note the expected binary figure |
| 256GB | 256,000,000,000 | ≈ 238.4 GiB | Same two-step reduction | Agree a byte floor and specify FAT32 or exFAT explicitly |
| 512GB | 512,000,000,000 | ≈ 476.8 GiB | Same two-step reduction | Add a file-size check where exFAT is required |
| 1TB | 1,000,000,000,000 | ≈ 931.3 GiB | Same two-step reduction (binary terabyte vs decimal) | Define the class in decimal bytes and name the acceptance host and OS |
Usable figures vary by tool and host, so treat the table as a reference for pre-agreeing tolerance rather than a fixed pass/fail number.
File system overhead: the second, smaller part of the gap
Even after the decimal-to-binary conversion, a freshly formatted volume shows a little less. A file system must build its own structures before storing user data: allocation tables, a directory and metadata area, and reserved or alignment regions. FAT32 and exFAT keep this cost modest, but it is never zero. The sequence is two-step — advertised bytes convert to a smaller binary figure, and formatting trims a further slice to leave the usable capacity. That second step is normal engineering behaviour, not a shortfall.
File system choice and the 4GB single-file limit
The file system is not a cosmetic setting; it decides the largest single file a customer can copy onto the product.
- FAT32 — max single file 4 GiB (4,294,967,295 bytes) and volumes up to roughly 2 TB; broadest device compatibility and a common default for smaller cards and basic-firmware hosts.
- exFAT — built from the outset for flash media; removes the practical 4GB single-file ceiling and works across modern Windows and macOS. The sensible choice for cards above 32GB and long video files.
- NTFS — journaling and Windows-first; extra write activity makes it a poor default for removable flash, and some non-Windows hosts mount it read-only.
The procurement rule: if a product is marketed for large video files, shipping it with FAT32 guarantees a predictable complaint.
How to specify capacity in a purchase order
- Name the capacity class in decimal bytes, not just the label. Write “64GB class = 64,000,000,000 bytes.”
- Agree a reporting tolerance up front — state the expected binary display (about 59.6 GiB for a 64GB class product) so the normal decimal/binary difference is pre-accepted and not rejectable.
- Specify the factory file system (FAT32 or exFAT) and any required volume label.
- Define the acceptance host and method. On a designated host and OS, verify raw capacity in bytes, then the formatted usable capacity at the file-system level.
- Add a file-size check. Where exFAT is required, copy a test file larger than 4 GiB to confirm the correct file system.
- Write the failure path. If reported bytes fall below the agreed byte floor, re-check and re-format before escalating; decimal/binary display gaps stay out of scope for rejection.
Capacity classes in the LingC storage range
The LingC portfolio spans the common capacity classes. The TFP3310 high-quality TF card covers 32GB to 256GB at up to 100MB/s; the TFU3320 high-speed TF card covers 64GB to 512GB at up to 250MB/s; the magnetic PS5005 portable SSD covers 512GB to 2TB at up to 2000MB/s. Warranty for the TF card and SSD/PSSD lines runs two years, final term on the quotation. Certification for the TF card models referenced here can be arranged upon order confirmation; the PS5005 carries CE/FCC/RoHS.
Shenzhen LingC Industrial Co., Ltd. operates a quality management system certified to IATF 16949. Explore the full range on the memory cards product line page.
FAQ
Is a 64GB TF card that shows about 59GB defective?
No. The label uses decimal capacity (1 GB = 1,000,000,000 bytes) while the OS reports in binary units, so a 64GB class product converts to roughly 59.6 GiB; file-system structures trim a little more. This is expected behaviour, which is why the tolerance belongs in the purchase order.
Should I order a card with FAT32 or exFAT?
It depends on the largest single file the end customer will handle and on host compatibility. FAT32 offers the widest compatibility but caps a single file at 4 GiB; exFAT removes that ceiling and suits cards above 32GB and video-heavy workflows. Specify the choice explicitly.
Why can’t I copy a file larger than 4GB onto the card?
The card is almost certainly formatted as FAT32, which limits a single file to 4 GiB (4,294,967,295 bytes). Re-formatting to exFAT removes the limit on modern hosts; if the product targets large files, exFAT should be the factory default.
What capacity tolerance should the purchase order state?
State the capacity class in decimal bytes, the expected binary display, and the acceptance host and method. Worked example: a 64GB class product should report about 64,000,000,000 bytes raw and approximately 59–60 GiB after formatting, with any shortfall below the agreed byte floor triggering a file-system re-check before escalation.
Summary
“64GB shows 59GB” comes from two sources: decimal marking meeting binary reporting, and a modest file-system overhead on top — neither indicates a faulty product. Once a buyer writes the decimal capacity class, the expected binary display, the factory file system and the acceptance method into the purchase order, that display gap becomes a controlled, pre-agreed specification.