There are 5 players: Endgame, Static, 5min, 20sec and Toronto.

Endgame: plays the static best as long as the bag isn't empty,
otherwise uses endgame solver.

Static: always plays the static best move.

The other three are really one player with different parameters
(Resolvent class).  The basic idea is to delegate to different methods
under different phases

Empty bag: use endgame player
1-2 tiles in bag: use preendgame player
otherwise: use "bogowin" player

The parameters are
5min - use 5 minutes per turn
20sec - use 20 sec per turn
Toronto - use 66 sec per turn

Each players tries to do what it can in the time given to play
according to its method.  I'm not sure if the details matter to you or
not - feel free to ask for more info.

Matt

Reply via email to