A tarot deck has more than only trumps. It also has fourteen non-trumps in each of four suits (Latin-suited or French-suited depending on the deck), so the total number of cards (trumps and non-trumps) is 78.

Furthermore, if you can store bytes, then why should it need to be UTF-8? Not all data is text, and not all text is Unicode (and not all Unicode text is most efficient with UTF-8). As far as I can tell, the only part of the program that requires it to be UTF-8 is the "decode_array_to_string" function (although I think it will have to be a string without embedded null characters since that is how argv is working, and that could possibly allow you to store a few more bytes, if the data is required to not have null characters).