Decoding UTF-16 (with surrogate pairs) is pretty similar to decoding UTF-8 except the state machine is simpler (as it's just 1–2 code units per codepoint, instead of 1–3). It also means that if you validate that you have a high surrogate and a low surrogate, the combination is guaranteed to be a valid codepoint (whereas with UTF-8 there are invalid encodings, e.g. any 1-byte codepoint can actually be encoded using 2 or 3 bytes instead, which the state machine must reject).

"A surrogate pair can represent 20 bits, or 2^20 values. A single UTF-16 byte is of course 16 bits, or 2^16 values. 2^20 + 2^16 == 0x110000, which is precisely as many unicode code points as there are (0–0x10FFFF)."

This explanation can't be quite right, because the possible surrogate code unit values are carved out of the set of 2^16 code units. So you are double counting those surrogate code units.

From Wikipedia: "Since the ranges for the high surrogates (0xD800–0xDBFF), low surrogates (0xDC00–0xDFFF), and valid BMP characters (0x0000–0xD7FF, 0xE000–0xFFFF) are disjoint, it is not possible for a surrogate to match a BMP character, or for two adjacent code units to look like a legal surrogate pair."

"The Unicode standard permanently reserves these code point values for UTF-16 encoding of the high and low surrogates, and they will never be assigned a character, so there should be no reason to encode them. The official Unicode standard says that no UTF forms, including UTF-16, can encode these code points."

Ugh.