Why Your Storage Never Quite Matches What You Bought: GB vs. GiB Explained

Buy a "1TB" hard drive, plug it in, and check its capacity in your file explorer — and it will very likely show something like 931GB, not 1,000GB. Nothing was returned, nothing is defective. Two different, both technically correct, ways of counting bytes are simply disagreeing with each other.

Decimal units: how storage is sold

Storage manufacturers measure capacity in decimal units, consistent with how every other metric measurement works: 1 kilobyte = 1,000 bytes, 1 megabyte = 1,000 kilobytes, 1 gigabyte = 1,000 megabytes, and so on in clean powers of 1,000. This is a defensible, standards-compliant way to measure — it's literally what "kilo," "mega," and "giga" mean in the International System of Units used across science and engineering.

Binary units: how your operating system counts

Computers, at the hardware level, work in powers of two, and early computing conventions carried that into how software reports storage: 1 kibibyte = 1,024 bytes, 1 mebibyte = 1,024 kibibytes, and so on in powers of 1,024. Many operating systems — Windows' file explorer among the most visible examples — report storage this way, while confusingly still labeling the units "KB," "MB," "GB" rather than the technically correct "KiB," "MiB," "GiB."

Why the gap grows at larger scales

The difference between 1,000 and 1,024 is about 2.4% at the kilobyte scale, but it compounds at each larger unit — by the terabyte scale, the gap between decimal and binary measurement widens to roughly 10%. That's why the discrepancy becomes genuinely noticeable — hundreds of gigabytes "missing" — on larger modern drives, even though the underlying percentage gap has been consistent all along.

Where each convention actually applies

RAM capacity is a genuinely binary quantity because of how memory chips are physically addressed — 8GB of RAM really is exactly 8 × 1,024³ bytes, no ambiguity involved. Storage drive capacity and network/internet speeds, on the other hand, are conventionally decimal, following the same metric-prefix logic as every other unit of measurement. Software that reports storage using binary math but decimal-style labels is the actual source of the confusion, not either measurement system being "wrong" on its own.

The practical takeaway

Nothing needs fixing when a drive shows less capacity than its label — it's simply being measured with a different, equally valid ruler. Understanding which convention applies to what you're looking at (decimal for what you bought, likely binary for what your OS displays) removes the mystery, and converting between the two directly shows exactly where the "missing" capacity actually went: nowhere, it was never missing to begin with.

A quick way to check the math yourself

To convert a decimal figure to its binary equivalent at the same scale, divide by roughly 1.0737 (the accumulated gap by the gigabyte level) — a 1,000GB decimal figure divided by that factor lands close to 931GiB, matching what a binary-reporting file explorer would show for the same physical drive. Running a few real numbers through the conversion, rather than just reading about the discrepancy, is usually what finally makes the two systems click as genuinely different measurements rather than one of them being an error.