AI Game Development: Synthetic Creatures with Learning and Reactive Behaviors (Paperback)

~ (Author)

Editorial Reviews

Product Description

Neural networks, decision trees, genetic classifiers: If these are AI concepts you'd like to employ in your own games-and you know your way around C++-this is the book for you! In these pages, leading game AI developer Alex J. Champandard shows you how to create a slew of autonomous synthetic creatures-in the process exploring the techniques and theories central to AI game development. Complex concepts are made easily graspable, even fun, as you apply them to the step-by-step development of your own complete bot. The focus here is on designing individual creatures, each with unique abilities and skills. Each chapter tackles a specific problem, using demos and examples to drive the points home. Best of all, Alex draws on his own real-life experiences to provide tips and tricks to speed the process and resolve thorny issues. On the companion Web site, you'll find code examples and the samples of some of the games covered in the book.

From the Back Cover

Learn how to add realistic AI characters to your games developed in C++ as well as commercial games, such as Quake 2 or Unreal. AI Game Development introduces a variety of AI techniques, such as neural networks, decision trees, genetic classifiers, and reinforcement learning. It also builds a complete bot step by step, providing practical demos and example code within an AI library freely available on the web site. Offering AI tips and tricks from one of the leading game AI researchers and developers AI Game Development outlines a variety of popular AI techniques applicable to game development, with novel insights into the design process. These techniques model characteristics from human intelligence, applied as algorithms that the computer can understand. The book is broken into sections, each offering detailed descriptions of the process of creatingautonomous synthetic creatures, beyond the level of intelligence of those found in current computer games. The focus is placed on the design of individual creatures, each with unique abilities and skills. Each chapter tackles a specific problem, supported by demos and examples coded within an open source game AI project named FEAR (fear.sourceforge/). Techniques used to solve each problem are described from a more generic point of view, so various theoretical concepts can be understood and successfully applied in different situations. The author also emphasizes key concepts that arise in the creation of any form of intelligence, and presents tips and tricks to deal with these issues. You will be equipped to design and experiment with AI techniques on their own,applying them to new problems as they see fit.

See all Editorial Reviews

Product Details

• Paperback: 768 pages
• Publisher: New Riders Games (December 6, 2003)
• Language: English
• ISBN-10: 1592730043
• ISBN-13: 978-1592730049
• Product Dimensions: 9.1 x 7.4 x 1.2 inches
• Shipping Weight: 2.8 pounds
• Average Customer Review: 3.4 out of 5 stars  See all reviews (8 customer reviews)
• Amazon.com Sales Rank: #1,103,160 in Books (See Bestsellers in Books)

Customer Reviews

Most Helpful Customer Reviews

18 of 23 people found the following review helpful:

5.0 out of 5 stars An excellent, well-presented look into AI, December 8, 2003

By	Dave (Omaha, NE) - See all my reviews

For so long, the computer gaming industry was concerned with making their products look and sound better - and for good reason. There have been many instructional books and articles that have paced the advances in graphics and sound technology. Game AI, however, has only recently come into the primary focus as the next frontier of game advancement and the resources available to AI programmers have been few and far between.

After spending time on game AI message boards, I have seen the endless torrent of people asking the simplest questions about game AI in general. I have also seen those who have a grasp of the basic tools but are unsure how to apply them in their situation. Also present are those who are looking for a better way to model larger problems but are faced with a wide array of academic, non-practical research and theories. "AI Game Development" caters to all of those people... from those that are learning the basics up through those that want to apply just that extra "nudge" to their projects. It fills a valuable need in the game development community to provide that "go to" tome to allow even the beginner to make steps in adding true interactive gameplay into his or her products.

So often people find a tool and look for a way to use it in their game. They fail to stand back and look at their problem and then select the correct tool, or combination of tools, for the job. Mr. Champandard has supplied the reader with reasonable examples and, more importantly, ways of thinking about and dissecting their problem so as to choose that correct tool.

Rather than a disjointed collection of tips and tricks, this book provides a smooth, overarching treatment of the concepts behind the AI. It explains why and what rather than how. Alex has a disciplined, outlined style that breaks down a very complex, multi-faceted subject into separate but connected stepping stones. Each one can be mentally processed alone, but always with the awareness that is a part of the whole series. The reader is not bombarded with pages of code, but is supplied with the reasons to make the code that way.

All in all, "AI Game Development" is an excellent piece of work from a very knowledgeable source and should be on the shelf of every game AI programmer. After all, as AI programmers, we ARE trying to build enemies, not targets!

23 of 30 people found the following review helpful:

1.0 out of 5 stars Poorly Written Original Content, May 26, 2004

By	David Ong (Australia) - See all my reviews

This book is the only book of it's kind with the Game AI content in the world at this point (late May 2004). So if youre interested in Game AI it's the only book you can get, however it is poorly written, has mistakes, typos, and insufficent description of the FEAR platform that is required for you to be familiar with if you want to understand the content of the book thoroughly. This view is maintained by my peers, tutors and lecturers in the University i go to which is currently using this book as a guide to our game devalopment unit (because its the only one of its kind).

I can't help but feel that the book was kinda rushed and was not properly/thoroughly proofed read by the editor. I am confident that a lower priced 2nd edition with all the mistakes taken out would be really appropriate and appreciated by game AI interested people. If not some one else should become the 2nd person in the world to write a book on this subject: Game AI.

The author should be a little self critical and make a 2nd edition and make us all happy instead of living in denial and giving him self 5 stars. Dont see how he can improve on him self without being self critical. We are all greatful here in my University for his initiative in inventing the first book on game ai but were also all very frustrated when strugling to asimilate it's content.

Also a great deal of the SOURCE CODE did NOT WORK. Some of them did, but a large fraction of them DID NOT work as expected from the book.

Granted some chapters that covered basic principles were easy to follow but some intermediate chapters really made me frustrated as i re-read them again and again to try and understand a concept but continue to fail unless i label certain sentences as mistakes that were worded wrongly or backward etc.

I was going to rate this book 2 stars because i felt that some chapters were well written and covered the content well, but since so many people already gave it an unrealistic score of 5 stars (perhaps a conspiracy by the author's friends) i have decided to compensate for this and lower my rating to 1 star.

Hopefully this will balance out the scoring to a fair and realistic level that it deserves.

David Ong, Australia

2 of 2 people found the following review helpful:

3.0 out of 5 stars Good if you already know AI and want to get into game development, January 2, 2006

By	Rodrigo Damazio (São Paulo, Brazil) - See all my reviews (REAL NAME)

This book is good in the sense that it sheds some light on how to apply many well-known algorithms (such as neural networks) in games, but it doesn't go very deep in any of those techniques, leaving the reader to understand many details only by reading the provided source code.

Some chapters, such as the ones on genetic algorithms, could have some improvement and more practical examples.

It is good as an introductory book for someone who already has notions of AI algorithms, though. For anyone looking for a deeper coverage of algorithms, I'd suggest the book "Artificial Intelligence: A Modern Approach" by Stuart Russel.