UTF-8 vs UTF-16: Encoding Size Comparison

UTF-8 vs UTF-16: Which Is Smaller?

The answer depends entirely on the content. UTF-8 and UTF-16 use different byte widths for different Unicode ranges, so the most efficient encoding varies by text language and composition.

Byte Widths by Character Range

Unicode Range	Characters	UTF-8	UTF-16
U+0000–U+007F	ASCII	1 byte	2 bytes
U+0080–U+07FF	Latin Extended, Greek, Cyrillic, Arabic, Hebrew	2 bytes	2 bytes
U+0800–U+FFFF	CJK, Devanagari, Thai, most symbols	3 bytes	2 bytes
U+10000–U+10FFFF	Emoji, rare CJK, historic scripts	4 bytes	4 bytes

When UTF-8 Wins

UTF-8 is more compact for text that is predominantly ASCII or Latin Extended:

"Hello World" (11 chars)
  UTF-8:  11 bytes
  UTF-16: 22 bytes  (2× larger!)

For English, French, German, Spanish, and other Latin-based languages, UTF-8 is typically 30-50% smaller than UTF-16.

When UTF-16 Wins

UTF-16 is more compact for text that is predominantly CJK:

"東京都渋谷区" (5 chars)
  UTF-8:  15 bytes
  UTF-16: 10 bytes  (33% smaller!)

Break-Even Point

For mixed-script text, the break-even depends on the ratio of ASCII to CJK characters. Roughly:

• More than 33% ASCII → UTF-8 wins
• Less than 33% ASCII → UTF-16 wins

Real-World Examples

Content Type	Typical UTF-8 / UTF-16 Ratio
English prose	0.5× (UTF-8 half the size)
JSON API response	0.6× (mostly ASCII keys + values)
Japanese blog post	1.2× (UTF-8 slightly larger)
Chinese document	1.4× (UTF-8 40% larger)
Emoji-heavy chat	1.0× (roughly equal)

Why UTF-8 Dominates the Web

Despite UTF-16 being smaller for some content, UTF-8 dominates because:

1. ASCII compatibility: No overhead for HTML tags, JSON syntax, URLs
2. No byte order: UTF-16 needs BOM (Byte Order Mark); UTF-8 does not
3. Self-synchronizing: You can find character boundaries from any byte position
4. Universal standard: HTML5, JSON, HTTP all specify UTF-8