Just finished skimming this wonderful book. I say I skimmed it since I read parts of it in detail, whereas I could only eye other parts of it. There were things in it that I knew before and I read them curiously. There were things in it that I knew before and I didn't read them thinking that my versions of concepts were well formed.
There were things in it that I did not know and I read them eagerly. There were things in it that I did not know before and I deferred reading them for a better day.
The writers have done a great job at explaining what goes in developing artificial intelligence (AI) for games. Algorithms for pathfinding, movement, decision making, and strategy are really profound and interesting. Why and when we need each of the feats is also explained pretty well.
However, the reader begins to put everything in context when he reaches the twelfth chapter. Here only realizes in a systematic way that what really goes in developing AI for games. Rest of the earlier stuff begins to make sense.
So the best way to read this book is to skim it up to the twelfth chapter in a first pass. Once you have reached there, you would have to make some connections with the dots. Dots are always connected backward, so start reading it back again, and you will find it quite awe-inspiring.
I call it wonderful because it indeed is quite wonderful. And quite insightful as well!