Twobytepercharacter

Twobytepercharacter is a term used to describe an encoding scheme that assigns exactly two bytes to each character. This fixed-width approach simplifies memory layout and indexing, since every character occupies the same amount of space. In practice, the most widely cited realization of this concept is UCS-2, an early Unicode encoding that stores each code point in 16 bits.

UCS-2 covers only the Basic Multilingual Plane, from U+0000 to U+FFFF, and therefore cannot represent characters from higher Unicode planes. To encode the full set of Unicode characters, UTF-16 uses 16-bit units but can combine two units as surrogate pairs, effectively allowing four bytes for a single character. Because of this, the blanket claim of two bytes per character does not hold for modern Unicode encodings.

Endianness affects two-byte encodings: the same two-byte value can represent different code points depending on whether the bytes are stored big-endian or little-endian. Some systems use a byte order mark to indicate the encoding order; others rely on explicit BE/LE conventions in the protocol.

Usage today is largely limited to legacy software and systems that need fixed-width, two-byte character fields for compatibility or performance reasons, or in data interchange with older protocols. For broad Unicode support, UTF-8, UTF-16, or UTF-32 are preferred, with the latter two offering fixed-width or semi-fixed-width options depending on the encoding.

Related topics include Unicode, UCS-2, UTF-16, surrogate pairs, and endianness.