HEDGE1'S PROFILE

OMG. That language selection option is really fancy. So are the books. I hope you're not putting too much work into all of these, though. Like, seriously, it's awesome and all, but writing 100 mini books to fill your bookcases and translating them all is really time consuming.

I approve of the animations, but I am saddened by the wide variance in damage output. I'm hoping that's either because you're just testing or because the weaker ones are from non-damage oriented individuals.

Concerning the possible slow down from a more robust AI script, I wouldn't worry so long as it stays within 1-2 seconds. Right now the instantaneous response is actually a bit jarring, and a short "thinking" time for the AI would make me feel like it's trying a bit harder, even if it isn't (and especially if it is). Bottom line: don't let it get too long, but a little wait for the AI turn is actually probably preferable, and if you can get the Monte Carlo to work in that time frame all the better. To save time, you could perhaps focus on eliminating options with some simple checks first, and if the move passes the simple checks then run the full Monte Carlo). Also, I agree with the idea of using a sub-set of possible cards (weighted based upon the player's "level") rather than a mirror of the AI's cards for the Monte Carlo.

All this AI talk is actually pretty exciting.

I agree with Dyhalto on the camera pans. The pace of the battle is fast enough that it feels a bit spastic and disorienting, and I definitely think you'd lose more by slowing down the battles than you would by ditching the camera pans. Plus, as Dyhalto also pointed out, you cut off some information with the camera movement. Otherwise it's a nice looking first reveal!