Brian Schwab (San Diego, CA) has been a game programmer for over ten years, and has held key positions as Gameplay and AI Programmer for both Angel Studios and DreamWorks Interactive. He currently works as a Senior AI Programmer for Sony Computer Entertainment.
The book is quite poorly edited. Images are stretched, blurry, oddly positioned and occasionally unlabeled. As a result of this, the size of the font in images fluctuates depending on where the image is positioned. In spite of the second edition being published in the end of 2008, their history of AI in video games ends in 2001, which makes me think that not much has changed in the book since the first edition. Pages are used up in unannotated pasted source code, which while interesting would have been much better if the core techniques and design philosophies were explained. Instead, it's just "here's some of the AI code from Game X" followed by five pages of line wrapped AI code from Game X. Not entirely worthless, but lazy. The later chapters covering neural networks and genetic algorithms are neat, but shallow.
The CD included with the book contains code for each of the chapters. The code, as far as I've looked at it is complete, and written to a decent level of quality. Strangely, there are no options to configure within the projects though to see the different approaches to AI - for example, different weights on various constraints, or different visualizations. Also strangely, the code is not available online, so don't lose the CD or you are out of luck. One thing that the book does well is provide many suggestions of ways to add functionality to the AI within the demos.
AI Game Engine Programming could have been well done, but the code demos need to be cleaned up and the book is in desperate need of a competent editor. As an alternative, I would recommend Programming Game AI By Example. It goes into much more depth, the code samples are far more advanced and complete, and the editing is anything but lazy.
This was a good book to cover AI Game Engines and gave the information at a fairly quick pace to keep you interested. A lot of topics are covered by this book, and I would definitely recommend picking it up if you are looking for an introduction to Game AI. This also covers scripting, plug-ins, and many topics that are required to write a Game AI Engine. Many different types of AI are also discussed. This will certainly stay on a shelf by my desk for reference.
5.0 out of 5 starsGood book for beginning AI programming
June 9, 2009 - Published on Amazon.com
Format: Paperback|Verified Purchase
This is a good book if you are a relatively experienced programmer who is beginning game programming (like me) or if you are already somewhat experienced in basic game development and simply want to improve your AI knowledge. The book covers loads of different AI patterns for all sorts of different situations. Some of these I was familiar with such as Finite State Machines but there were plenty of others that I hadn't taken a look at before but which am now using a lot (thanks to this book) such as Fuzzy State Machines. I didn't really have any trouble with the graphics code being written in OpenGL which is my library of choice since I do a lot of work on non-Microsoft systems. If you are also in the process of learning OpenGL, for example because you want to write games software for mobile platforms such as the iPhone/iPod or Android, this is actually a nice bonus. Mind you the Apple products at least use OpenGL ES which differs slightly from OpenGL but this doesn't detract from the value of the sample OpenGL code provided here. I like the way the author demonstrates his teachings with quite comprehensive example projects unlike some other programming books I have read where the authors provided a lot less code. In the end there are few things that you learn from as well as reading raw sources. Quite a lot of the source code is written in C++ but then again I didn't buy the book for C++ lessons, I bought it for the design patterns it covers and these are easily implemented in other languages. The one criticism I have of this book is that after only a few months of use the binding is already splitting.
This book is awful, i have read this as well as the earlier book by the same name from this author. This book is 75 percent completely useless filler, not to mention the code examples are overflowing with unimaginable bugs (functions definitions with no return types, basic logic and syntax errors and much worse), obviously they did not bother even trying to compile any of them and you will spend a significant amount of time fixing all these issues. This is not worth a purchase at any price greater than 3 dollars, you are better of searching google for information from various websites if you need a book any alternative is better.
3.0 out of 5 starsNo real meat and potatoes, but many appetizers.
August 22, 2005 - Published on Amazon.com
First, I would like to address the comment a reviewer made about the book being no good because it utilizes OpenGL.
This is a book on AI, the AI concepts and code can be implemented using any rendering API (or none at all) to visualize their activity. OpenGL and DirectX have as much to do with AI as car engine maintenance has to do with cooking a good lobster neuberg. The fact that the reviewer cannot make this distinction is sadly telling. Some people want all the work done for them. If you are that type, don't bother with programming.
Second, the reviewer who said that this book is too scattered is right. It tries to do too many things and ends up doing none of them extremely well. It does, however, have a few good code samples and can be used as an introductory book on the subject.
well as i skimmed through the book it was all good. i wish there was more examples. i haven't checked out the cd yet but other than that i still love the book. just one thing though. when i got mines it sort of had a big torn on the book so it looked like two different books flapping together. other than that the book was still in mint condition.