>>22937075well, a byte is 8 bits, and a bit (binary digit) is a digit with value 0 or 1, or 2 possible states
so a byte can represent 256 possible unique states (0000 to 1111), the same way a 4-digit combination lock can represent 10,000 possible unique states (0000 to 9999).
256 unique values is enough to contain all lowercase and uppercase letters, numbers, and some special characters.
this set of characters is called ascii.
a single ascii character, a single letter in my post, takes up 1 byte
a unicode character is different, it takes up 2 bytes.
2 bytes can represent one of 65,536 possible unique values (00000000 to 11111111), the same way an 8-digit combination lock can represent one of 100,000,000 possible unique values
65,536 different values is enough to contain basically every character from every language, which is what unicode is for.
the first 256 characters of unicode are ascii, meaning it contains the english alphabet as well as all sorts of kanji and cyrillic and shit.
i think filenames can be a max of 120 ascii characters, which is why an image with no filesize still takes up 120 bytes.