New Iagno AI

[Arnaud Bonatti <arnaud bonatti gmail com>]

Dear people interested in GNOME games,

As you may know, I maintain GNOME Reversi, aka Iagno. This day I’m
proud to announce it gained a new AI [for medium and hard levels].
Nothing completely new for real, that’s “just” tweaking of various
parameters, code refactoring, and various fixes done by understanding
deeply how the AI works; no new strategy coded or things like that.
Anyway, the result is impressive: the new AI is a depth 3, while the
previous one was a depth 6; it feels instant because of that, but also
because of its improved alpha-beta pruning, and of the various code
refactorings that happened before the 3.33.2 release; and even with
this important depth gap, it feels harder.

Here comes the reason of this mail: if you’re a
Reversi/reversi/Othello player, please, test and comment the new (git
master branch, or future 3.33.3) medium/hard AI behavior. It should
play correctly, offering some fun challenge. As previously, the two
first moves are done randomly, but it has (normally) no random
elsewhere, so if you find a winning strategy for one of the openings,
that’s a bug and it should be fixed.

Regards,
Arnaud

-- 
Arnaud Bonatti
________________________________
courriel : arnaud bonatti gmail com