The table below shows the categories of two-byte characters in the Shift-JIS character set. The range values are in hexadecimal. The second byte in a Shift-JIS character may have values between hexadecimal 40 and FC, excluding 7F.
Range
Type of Character
8140-81FC
Symbols
8240-824E
undefined
824F-8258
Digits
8259-825F
undefined
8260-8278
Uppercase Roman letters
8279-8280
undefined
8281-8299
Lowercase Roman letters
829B-829E
undefined
829F-82F1
Hiragana
82F2-82FC
undefined
8300-8396
Katakana
8397-839E
undefined
839F-83B6
Uppercase Greek letters
83B7-83BE
undefined
83BF-83D6
Lowercase Greek letters
83D7-83FC
undefined
8440-8461
Uppercase Russian Cyrillic
8462
undefined
8463-8491
Lowercase Russian Cyrillic
8492-849E
undefined
849F-84BE
Box drawing
84BF-84FC
undefined
85xx-86xx
undefined
8740-879C
symbols
879D-97FC
undefined
8840-889E
undefined
889F-8FFC
Level 1 kanji
89xx-9872
Level 1 kanji
9873-989E
undefined
989F-9FFC
Level 2 kanji
E0xx-EFxx
Level 2 kanji
F0xx-FCxx
Level 3 kanji (undefined in most implementations)
The level 1 kanji are arranged in the order of the hiragana representations of their most common on pronunciation. The level 2 and level 3 kanji are arranged in order of the stroke count of their principal radical followed by the stroke count of the remaining portion of the character.
All two-byte characters, including the Roman, Greek, and Russian characters, are twice as wide when displayed as the one-byte characters. The HTBasic CVT$ function can convert between one- two-byte Roman and katakana characters.
Note that voiced katakana and hiragana characters are represented by a single two-byte character in the shift-JIS character set while they are represented by a one-byte character plus a separate one-byte voicing mark in the ISO-932 character set.
