Game Development Reference
In-Depth Information
early and another move that would reform their line later. The early move would
have more black squares, thus a lower board rank, so the hounds would pursue
the later move. ''Instead of slamming the door in your face now, I'll lead you on
and just do it to you later.''
The AI.V5.vb file on the CD has the code for this method. The routines of interest are the better
move routines in the ''Internal Stuff'' region. Use it with the na¨ve method for counting black
squares (mentioned later in this chapter) to get the behaviors shown in Figures 6.7 through 6.9.
The hounds can do this if the fox moves the way the hounds expect the fox to
move. In games where the fox AI is pitted against the hounds AI, this is always
true. Recall that when the hounds are looking ahead, it is to fix their broken line.
This is the time that the fox is not looking ahead, but instead taking the shortest
path to freedom. The fox is not required to play the way the hounds want it to. A
few moves of human intervention helps the fox by ignoring a path to freedom
that the hounds will close before the fox escapes. Instead, the fox blocks the
hounds from making critical moves on which their plan depends.
This is illustrated in the next few figures. The boards shown are after the hounds
make a move, hence the even move numbers. The action starts in Figure 6.7 with
Figure 6.7
The hounds break their line and plan for a glorious future.
Search Nedrilad ::

Custom Search