Unit Converter Tools

Bits, Bytes, KB, MB, GB, TB Explained - The Complete Data Storage Guide

EveryTool Editorial
7 min read

Data storage is measured in units that most people use every day but few fully understand. Why does your 1 TB hard drive show only 931 GB in Windows? Why is a 100 Mbps internet connection so much slower than it sounds when downloading files? Why do some operating systems show "GB" while technical documentation says "GiB"? These questions all have precise answers rooted in the history of how computers were designed and how the storage industry evolved alongside them. This guide explains all of it clearly.

The Fundamentals - Bits and Bytes

The bit (binary digit) is the most fundamental unit of digital information - a single value that is either 0 or 1. All digital data - text, images, video, code, audio - is ultimately stored as sequences of bits. Eight bits make one byte, which is the practical base unit for measuring file sizes and storage capacity. The byte became the standard because it is large enough to represent a single character of text (in ASCII encoding) while remaining a manageable unit for calculation. From the byte all larger units are derived by multiplying by either powers of 1,000 (decimal) or powers of 1,024 (binary) - and this is where the confusion begins.

Two Standards - Decimal and Binary

The core confusion in data storage measurements comes from two competing standards that use the same unit names for different values. The decimal (SI) standard, used by storage manufacturers, defines each prefix as a power of 1,000: 1 kilobyte = 1,000 bytes, 1 megabyte = 1,000,000 bytes, 1 gigabyte = 1,000,000,000 bytes. The binary standard, rooted in how computers actually address memory, defines each prefix as a power of 1,024: 1 kilobyte = 1,024 bytes, 1 megabyte = 1,048,576 bytes, 1 gigabyte = 1,073,741,824 bytes. For small units the difference is minor - kilobytes differ by only 2.4%. But by the time you reach terabytes the gap is nearly 10%.

Quick reference: 1 GB (decimal) = 1,000,000,000 bytes. 1 GiB (binary) = 1,073,741,824 bytes. The difference is about 7.4%. By 1 TB the gap grows to about 9.1%.

Why Your Hard Drive Shows Less Space Than Advertised

A hard drive manufacturer labels a drive as "1 TB" meaning 1,000,000,000,000 bytes in the decimal standard. Your operating system - whether Windows, macOS, or Linux - reads those same bytes using the binary standard and displays the result in what it calls "GB." Since 1,000,000,000,000 bytes ÷ 1,073,741,824 bytes-per-GiB = 931.32 GiB, your OS shows approximately 931 GB. Not a single byte is missing. The drive contains exactly what was advertised. The discrepancy is purely a measurement standard conflict. This issue grows with drive size - a 4 TB drive shows about 3.64 TB in Windows.

The IEC Solution - KiB, MiB, GiB

In 1998 the International Electrotechnical Commission (IEC) introduced new prefixes specifically to resolve this ambiguity. They defined kibibyte (KiB) = 1,024 bytes, mebibyte (MiB) = 1,048,576 bytes, gibibyte (GiB) = 1,073,741,824 bytes, and so on up to yobibyte (YiB). Under this standard kilobyte (KB) unambiguously means 1,000 bytes and kibibyte (KiB) unambiguously means 1,024 bytes. Technical standards bodies, Linux kernel documentation, and many developer tools have adopted this notation. However storage manufacturers still use KB/MB/GB to mean decimal values, and Windows still labels binary values as GB rather than GiB - so the confusion persists in practice.

Bits vs Bytes in Networking - A Critical Distinction

  • Internet and network speeds are measured in bits per second - Mbps (megabits per second) and Gbps (gigabits per second).
  • File sizes and storage are measured in bytes - MB, GB, TB.
  • Since 1 byte = 8 bits you must divide a speed in Mbps by 8 to get the file transfer rate in MB/s.
  • A 100 Mbps internet connection transfers approximately 12.5 MB per second - not 100 MB/s.
  • A gigabit (1 Gbps) ethernet connection transfers approximately 125 MB per second.
  • ISPs advertise speeds in Mbps specifically because the numbers are 8× larger than the MB/s rate users actually experience.

When an ISP advertises '100 Mbps', that is megabits not megabytes. Your actual download speed in the units your download manager shows (MB/s) will be about 12.5 MB/s - divide the Mbps number by 8.

The Full Scale - From Bits to Yottabytes

  • Bit (b) - a single 0 or 1
  • Byte (B) - 8 bits - enough for one ASCII character
  • Kilobyte (KB) - 1,000 bytes (decimal) / 1,024 bytes (binary) - a short text file
  • Megabyte (MB) - 1,000,000 bytes - a compressed photo
  • Gigabyte (GB) - 1,000,000,000 bytes - a movie or a few hundred photos
  • Terabyte (TB) - 1,000,000,000,000 bytes - a typical laptop hard drive
  • Petabyte (PB) - 1,000 TB - large data centers measure in petabytes
  • Exabyte (EB) - 1,000 PB - global internet traffic is measured in exabytes per month
  • Zettabyte (ZB) - 1,000 EB - total global data storage is estimated in zettabytes
  • Yottabyte (YB) - 1,000 ZB - theoretical maximum of current naming conventions

Frequently Asked Questions

Is a gigabyte 1,000 or 1,024 megabytes?

Both - depending on the standard. Decimal (used by storage manufacturers): 1 GB = 1,000 MB. Binary (used by operating systems): 1 GiB = 1,024 MiB. The IEC recommends using GiB for the binary value and GB for the decimal value.

Why do some files show different sizes in different programs?

Different programs use different standards. A file manager might show a file as 1.4 GB (binary) while the same file shows as 1.5 GB (decimal) in another tool. Neither is wrong - they are just using different conventions.

What is the largest unit of data storage?

Currently named units go up to yottabyte (10²⁴ bytes). Beyond that the terms brontobyte and geopbyte are sometimes used informally but are not official SI prefixes. The total estimated digital data in existence is measured in the low single-digit zettabytes.

How much data is a zettabyte?

One zettabyte is 1,000,000,000,000,000,000,000 bytes (10²¹ bytes) or 1,000 exabytes. The total amount of data created, captured, and replicated globally has been estimated at multiple zettabytes and growing rapidly each year.

Why are RAM sizes always in powers of 2 (4GB, 8GB, 16GB)?

Computer memory addressing is binary - memory chips are organized in rows and columns that are powers of 2. This makes binary multiples (1,024, 2,048, 4,096) the natural sizes. This is also why operating systems use binary measurement - they are counting actual memory addresses.

Tools Mentioned in this Article