Big5 and GB2312: A Brief History of Chinese Character Encoding

The challenge of encoding thousands of Chinese characters in a digital format predates Unicode by decades. The two dominant legacy encodings — Big5 (used in Taiwan and Hong Kong) and GB2312 (used in mainland China) — reflect not just technical differences but a political and cultural divide over simplified versus traditional Chinese characters.

GB2312

GB2312 (Guojia Biaozhun, 国家标准) was published in 1981 by the People's Republic of China. It uses a two-byte encoding covering 6,763 simplified Chinese characters plus Latin, Greek, Cyrillic, Japanese kana, and various symbols — 7,445 characters in total. The character set reflects the PRC's policy of using simplified characters. GBK (1993) and GB18030 (2000) are backward-compatible extensions that significantly expanded the character set and eventually provided full Unicode coverage.

Big5

Big5 was developed in Taiwan in 1984 by a consortium of five companies. It covers 13,060 traditional Chinese characters in a two-byte encoding. The "Big" in Big5 refers to the encoding being designed for "Big 5" corporations; the 5 refers to the five founding companies. Big5 has several vendor-specific extensions (HKSCS for Hong Kong, various Microsoft extensions) that add characters not in the original standard.

Technical Characteristics

Both are variable-width encodings where single bytes (0x00–0x7F) encode ASCII characters and double bytes encode Chinese characters. Like Shift-JIS, they faced the challenge of avoiding ASCII byte values in multi-byte sequences. GB2312 uses lead bytes 0xA1–0xF7; Big5 uses 0x81–0xFE. This makes them incompatible with each other.

Transition to Unicode

Both encodings have been largely superseded by UTF-8 for web content. GB18030 became a mandatory standard in China for software products, requiring full Unicode coverage. You can compare Big5 and GB2312 byte sequences with Unicode representations using our character browser.