Is 1 GB equal to 1000 MB or 1024 MB?

Both, depending on which standard. SI (decimal) says 1 GB = 1,000 MB = 1,000,000,000 bytes. IEC (binary) says 1 GiB = 1024 MiB = 1,073,741,824 bytes. The naming is what disambiguates: GB = decimal, GiB = binary.

Why do hard drives "lose" capacity?

They don’t. A 1 TB drive really has 1,000,000,000,000 bytes (decimal TB). Windows displays the same drive as 931 GB because it’s actually showing GiB but labeling it GB. macOS Finder switched to decimal in 2009; Windows still uses binary. The bytes are all there; the rulers are different.

Does it matter for everyday users?

Sort of. The discrepancy is small at the kilobyte level (2.4%) but compounds — about 7% at the gigabyte level, 10% at the terabyte. So a "1 TB" SSD that displays as 931 GiB isn’t broken; it’s the same bytes shown two ways. For network speeds and file sizes, the gap explains a lot of "where did my space go" confusion.

Why do RAM and CPU caches use binary prefixes?

Hardware addressing is binary. RAM is built in powers of 2 because each address bit doubles the addressable space. So 16 GB of RAM is genuinely 16 GiB = 17,179,869,184 bytes. Hard drives don’t have this constraint — they can be any size — so manufacturers use the friendlier round decimal numbers.

GB vs GiB — Settled. Why Your "1 TB" Drive Shows 931 GB

You bought a 1 TB hard drive. Windows says it’s 931 GB. Apple says it’s 1 TB. Linux says 931 GiB. None of them are wrong. They’re measuring the same bytes with two different rulers.

The two scales

SI / decimal — KB, MB, GB, TB. Each step ×1000. Used by hard drive manufacturers, ISPs, network speeds, and Apple Finder since 2009.
IEC / binary — KiB, MiB, GiB, TiB. Each step ×1024. Used by RAM specs, Linux/BSD utilities, Windows file properties, the “technically correct” crowd.

Why two scales exist

Originally, “kilo” in “kilobyte” was a casual borrowing — close enough to 1000 to use the prefix, but actually 1024 because computers are binary (2^10 = 1024). For decades, KB meant 1024 bytes everywhere in computing. Then storage marketing started using the strict SI meaning (1000 bytes) because it makes drive capacities sound bigger. The IEC introduced KiB/MiB/GiB in 1998 to disambiguate. Adoption is still uneven.

The gap compounds

1 KB vs 1 KiB — 2.4% difference (1024 / 1000)
1 MB vs 1 MiB — 4.9%
1 GB vs 1 GiB — 7.4%
1 TB vs 1 TiB — 10.0% (and growing — your “1 TB” drive shows as 931 GiB in Windows)

// TRY THE TOOL

CONVERT BOTH WAYS.

Drop a value in either scale and see the equivalent in the other. KB ↔ KiB, MB ↔ MiB, GB ↔ GiB, side by side.

OPEN →

Bits vs bytes — a separate confusion

Network speeds are quoted in bits per second (Mbps, Gbps). File sizes are in bytes. To translate: 1 byte = 8 bits, so divide bps by 8 to get B/s. A 100 Mbps connection downloads at about 12.5 MB/s, not 100 MB/s. ISPs love bps because the number is bigger.

What to use when

Talking about hardware capacity — use what the spec sheet uses (decimal for drives, binary for RAM).
Writing software — be explicit. 1024 * 1024 for binary; 1_000_000 for decimal. Comment the choice.
Talking to humans — use the scale the user’s OS shows them. Confusing them with KiB will not help.

The takeaway

A 1 TB drive has 1 trillion bytes — full stop. Whether your OS shows 1000 GB or 931 GiB is a labeling choice. Both are correct. The bytes are all there.

GB vs GiB. Settled.