Number bases, text and numerals
The same quantity written a different way. Every page shows the conversion being done, not just its result: a base is a positional sum you can check, and Roman numerals are a set of rules that reject far more strings than they accept.
The working, not the answer
1011 in binary is 1×2³ + 0×2² + 1×2¹ + 1×2⁰. Going back is repeated division, reading the remainders upwards. Both are printed.
UTF-8, stated plainly
Text becomes bytes, not characters. An accented letter is two bytes, so five characters can make six groups, and the page says so instead of quietly mangling it.
It refuses what is wrong
A 2 in a binary number is rejected rather than truncated. IIX reads as ten by naive subtraction and is not a numeral, so it is refused too.
Between bases
Text and bytes
Numerals and words
Arithmetic instead?
The calculator evaluates an expression and prints each operation in the order precedence applies it.