Multibyte

Multibyte is a term used in computing to describe character encodings that represent a single character with more than one byte. It contrasts with single-byte encodings such as ASCII, where each character is encoded in exactly one byte. Multibyte encodings are typically used to support large character sets such as the Unicode repertoire, including many languages with thousands of symbols.

The most common multibyte encoding today is UTF-8, which encodes the first 128 ASCII characters in one byte and uses sequences of two to four bytes for other characters. UTF-8 is backward compatible with ASCII and is widely used in data storage, networking, and applications. Other multibyte encodings include UTF-16, which uses 2-byte code units and can require a surrogate pair for some characters, and legacy encodings such as certain East Asian code pages (Shift JIS, GB2312, Big5) that mix one- and two-byte sequences. In Windows, the term MBCS (multibyte character set) is sometimes used to describe code pages that use both single- and double-byte characters.

Handling multibyte text introduces complexities. The number of bytes in a string does not equal the number of characters, and character boundaries must be detected when slicing, indexing, or performing I/O. C and C++ libraries expose functions to process multibyte sequences, such as mblen, mbrtowc, and mbstowcs, which rely on a per-thread state object. In Windows, APIs like MultiByteToWideChar convert between multibyte and wide-character representations. Correct normalization and locale settings are important to interpret bytes properly.

From a design perspective, UTF-8 has become a de facto standard because it handles all Unicode code points with compatibility for ASCII and without fixed byte width, enabling robust text processing across languages. When interoperability is required with older systems, multibyte encodings may still be encountered, making awareness of encoding boundaries and conversion essential.