Bytes and Data Size Converter

Binary versus decimal prefixes

The confusion around data sizes stems from two competing prefix systems. Hard drive manufacturers use decimal prefixes where 1 KB = 1,000 bytes, 1 MB = 1,000,000 bytes, and 1 TB = 1,000,000,000,000 bytes. Operating systems use binary prefixes where 1 KB = 1,024 bytes (2¹⁰), 1 MB = 1,048,576 bytes (2²⁰), and 1 TB = 1,099,511,627,776 bytes (2⁴⁰).

This is why a 1 TB hard drive shows as approximately 931 GB in Windows or macOS. The drive actually contains 1,000,000,000,000 bytes. Windows divides by 1,024 three times (for GB), arriving at about 931. Neither number is wrong; they use different definitions of "GB."

The IEC introduced unambiguous prefixes in 1998: kibibyte (KiB = 1,024 bytes), mebibyte (MiB = 1,048,576 bytes), gibibyte (GiB), tebibyte (TiB), and so on. Linux kernel documentation, some file systems, and technical standards use these binary prefixes to avoid ambiguity. Consumer software has been slower to adopt them.

This converter uses binary prefixes (1 KB = 1,024 bytes) to match what operating systems report, which is the most common use case when checking file sizes, RAM, or storage capacity in a system context.

Data sizes in practice

API request and response sizes are often constrained by payload limits. AWS API Gateway has a 10 MB payload limit. Many mobile APIs target payloads under 100 KB to stay within a single TCP window. Knowing how many bytes a JSON object occupies, or whether a base64-encoded image fits within a limit, requires quick size arithmetic.

Bandwidth is often quoted in bits per second (bps), not bytes. A 100 Mbps network connection transfers 100,000,000 bits per second, which is 12.5 MB per second (dividing by 8 because 1 byte = 8 bits). Confusing megabits with megabytes is a common mistake when estimating download times or streaming requirements.

Frequently asked questions