In an early version, there was a “back attack” that you could do by hitting back-and-attack simultaneously. In reality, this either meant I had to delay all attacks in order to see if the user was about to hit “back”, which felt basically like ass, or it meant that you had to hit “back” before doing the attack which was starting to give me hand cramps from testing.

In Nanok, one of the problems I ran into was that the enemies would attack too much. You really need for the enemies to back off often to give the player a chance to be awesome. I think, from some more thought, this has to be proportional to how many enemies are nearby – the more enemies there are, the more space they have to give the player – otherwise groups of enemies aren’t fun.

It’s always kind of sad when you have to program enemies to be intentionally stupid.

I’m not sure how to make the up-attack better, aside from giving the enemies a stronger but longer-delay down attack so you can actually respond properly. I’m also a little hesitant to make the enemies overcomplicated, but I suppose this is also why games tend to have many relatively-simple enemies that can be mixed up in interesting ways.

Things to think about.

Attacking while slowing down to turn performs a long attack in the direction currently facing. That doesn’t feel right.

It is difficult to hit anything with the straight-up attack, and the attack patterns almost guarantee that you have to take a down-hit to be in the situation where you can hit something.

I found it notable that the AI would sometimes fall back.