> The most novel aspect of OoT bitsets is that the first 4 bits in the 16-bit coordinate IDs index which bit is set. For example, 0xA5 shows that the 5th bit is set in the 10th word of the array of 16-bit integers. This only works in the 16-bit representation! 32 bit words would need 5 bits to index the bit, which wouldn't map cleanly to a nibble for debugging.
There is nothing novel about this really. It's neat that it works with hexadecimal printing to directly read off the sub-limb index but honestly who cares about that.
Outside of that observation there's no advantage to 16-bit limbs and this is just a bog-standard bitset where the first k bits indicate the position within the 2^k bit limb, and the remaining bits give you the limb index.
My comment was worded a bit too harshly, sorry about that, I certainly wouldn't have worded that way in a personal message.
As I mentioned the hexadecimal printing coincidence is a neat fact, I was just excited when clicking the link to find a novel bitset idea. In my disappointment to find the standard bitset (albeit with 16-bit limbs) I reacted a bit too harshly. And as per https://xkcd.com/1053/, just because something isn't new to me doesn't mean it's not new to anyone.
There is nothing novel about this really. It's neat that it works with hexadecimal printing to directly read off the sub-limb index but honestly who cares about that.
Outside of that observation there's no advantage to 16-bit limbs and this is just a bog-standard bitset where the first k bits indicate the position within the 2^k bit limb, and the remaining bits give you the limb index.