The one-line calculation
1 TB = 1,000,000,000,000 bytes. 1 GiB = 1,073,741,824 bytes. Divide and you get 931.32 GiB. Every drive shrinks by the same 7.38% when its capacity is expressed in binary units: 500 GB → 465.66 GiB, 2 TB → 1,862.65 GiB, 4 TB → 3,725.29 GiB.
Who uses which convention
- Drive, SSD, USB and memory-card vendors: decimal SI (1 TB = 1,000 GB), used by every manufacturer.
- Windows: binary arithmetic, but labelled “GB” instead of GiB — the source of the confusion.
- macOS since 2009: decimal, so Finder reports 1 TB for the same drive.
- Linux: depends on the tool — ls -H uses decimal, ls -h binary.
- Cloud providers: decimal for billing and quotas, often binary in their desktop clients.
What is genuinely reserved
After the unit conversion there are two small real deductions. SSDs keep 1–7% of the flash for over-provisioning, which never appears as usable space. Prebuilt machines sometimes hide a recovery partition of 10–20 GB. Both are visible in Disk Management or diskutil list — neither explains the missing 69 GB on a 1 TB drive.
How to verify it yourself
- Right-click the drive in Windows → Properties: the byte figure is the vendor's number (a 1 TB drive shows slightly over 1,000,000,000,000 bytes).
- Divide that byte count by 1,073,741,824 to reproduce the GiB the system reports.
- On macOS, Get Info in Finder shows decimal GB — the same bytes, a friendlier label.
- Cross-check with the calculator on this page: enter the advertised size and read both figures.
Why the industry never fixed it
The IEC introduced KiB/MiB/GiB in 1998 so that “kilobyte” could mean 1,000 bytes everywhere. Software had already spent two decades using 1,024, and renaming units inside operating systems risks breaking scripts, documentation and expectations. So the industry settled on a truce: hardware speaks SI, most software speaks binary, and pages like this one explain the gap.