Home › Data Size Converter

Data Size Converter

UtilityRuns in your browserNothing uploaded

Convert data sizes

The 1,000 versus 1,024 problem

There are two systems in circulation, and the disagreement between them causes constant confusion.

Decimal (SI) units use powers of 1,000. A kilobyte is 1,000 bytes, a megabyte 1,000,000, a gigabyte 1,000,000,000. This is what the SI prefixes have always meant — kilo means a thousand.

Binary (IEC) units use powers of 1,024, because computers address memory in powers of two. A kibibyte (KiB) is 1,024 bytes, a mebibyte (MiB) is 1,048,576, a gibibyte (GiB) is 1,073,741,824.

The IEC introduced the -bi- names in 1998 specifically to resolve the ambiguity. Adoption has been partial, which is why the confusion persists.

Why your drive shows less than the box says

This is the most familiar consequence. Drive manufacturers use decimal units, so a “1 TB” drive holds 1,000,000,000,000 bytes.

Windows reports capacity in binary units but labels them with decimal names, so it divides by 1,099,511,627,776 and displays 931 GB. Nothing is missing — the same bytes are being described using a different definition.

The discrepancy grows with size: about 2.4% at kilobytes, 4.9% at megabytes, 7.4% at gigabytes, and 10% at terabytes.

macOS and most Linux distributions switched to decimal units, so they report the drive as 1 TB and match the packaging. Windows retains the older convention.

Bits versus bytes in networking

A separate and equally common confusion. Storage is measured in bytes; network speed is measured in bits. There are 8 bits to a byte.

A 100 Mbps connection therefore delivers a theoretical maximum of 12.5 MB per second, not 100. The abbreviations differ by one letter — Mb is megabits, MB is megabytes — which is why the distinction is missed so often.

In practice, real throughput runs 10–20% below the theoretical figure because of protocol overhead: TCP headers, IP headers, and acknowledgements all consume bandwidth without carrying payload.

Rough sizes for context

Where it actually matters

Cloud billing. Providers differ in whether they charge per GB or GiB, and at scale the 7.4% gap is real money. Check the documentation rather than assuming.

API and upload limits. A stated 10 MB limit may mean 10,000,000 or 10,485,760 bytes. Files near the boundary will fail unpredictably if you guess wrong.

Memory allocation. RAM is always sold and addressed in binary units, since memory addressing is inherently binary. An 8 GB module is 8 GiB.

Capacity planning. Mixing the two systems across a large estimate compounds the error, which is how storage projections end up 10% short.

Frequently asked questions

Why does my 1 TB drive show as 931 GB?

The manufacturer uses decimal units where 1 TB is a trillion bytes. Windows divides by 1,099,511,627,776 but labels the result GB. The same bytes are present - only the definition differs.

What is the difference between MB and MiB?

A megabyte (MB) is 1,000,000 bytes in decimal units. A mebibyte (MiB) is 1,048,576 bytes, or 2 to the power 20. The IEC introduced the binary names in 1998 to remove the ambiguity.

Why is my 100 Mbps connection only downloading at 12 MB/s?

Network speeds are measured in bits and file sizes in bytes, with 8 bits to a byte. 100 Mbps is a theoretical 12.5 MB/s, and real throughput is typically 10-20% lower due to protocol overhead.

Is RAM measured in decimal or binary units?

Binary. Memory addressing is inherently based on powers of two, so an 8 GB module is 8 GiB - 8,589,934,592 bytes.