Game Coding Complete [Paperback]

Chapter 1, Game Programming is Wacky starts the book off really well, distinguishing game programming from any other kind of programming. The rest of Part I is a gentle introduction into the high level decisions a programmer has to make before they write their first line of code.

Part II is my favorite part of the book because no other game programming book I've ever read, and I've read a few, exposes important topics like multiprocessing, resource caching, or why the stock ANSI random number generator isn't your best choice.

Part III continues this trend, digging into architectural topics such as 3D worlds and logistical topics like debugging games. Most game programming books show you how to write code, this is the first one (that I know of) that shows programmers how to fix it when it is broken. The section on special considerations for Windows games is especially useful for any Windows programmer who writes games or real time simulations.

Part IV takes a break from programming and discusses some production topics like scheduling and testing. Programmers who know this stuff will always finish their work on time and enjoy their weekends away from the keyboard.

Enough niceties - here's some harsh commentary from the author about the author.

First and foremost, the source code in the book is completely broken and disorganized. If someone spent the hours of typing necessary to write each line of code into a single compiling project they'd be rewarded with hundreds of compiler and linker errors.

Thank heavens for the web site, and the ability to download code that actually works.

Next, there are some serious omissions in this book.

Game audio is an important subject, and tends to be ignored until the last minute in most games. There should be a chapter that discusses DirectSound, Miles Audio, and how to synchronize sounds and animations. Character speech would be a good topic to add as well. The fact that the debugging chapter makes such a big deal of audio bugs makes this missing chapter even more apparent.

Game physics and collision is also totally missing, except for a brief mention of Havok in the 3D engines chapter. Physics and collision are critical to every game project, even something as simple as breakout. The author should be drawn and quartered for this omission.

Finally, there's absolutely no mention of networking and multiplayer game technolgy. The fact that the author has worked on some of the most famous (or notorious) multiplayer and massively multiplayer games makes this omission even more bizarre. Any book from Mr.Mike should have addressed playing games over the internet.

Enough complaints - here are some excuses:

First, the book was supposed to be no more than 550 pages. You'll notice that the book weighs in at more than 580. What you don't know is how many chapters were cut completely, or cut for length. The user interface chapter could have been twice as long, continuing the discussion with implementing various kinds of controls for games. There were plans for a physics chapter, and a networking chapter - they were cut because the book was running long and the deadline was nigh.

My only solace is that some game programming books are 800+ pages long, and say a lot less than you could have found in the DirectX help. If this book does well, I promise to include more info on audio, collision, physics, and networking in the next revision.

All said, I'm very happy with this book - even some of my beta readers told me they learned things when they read the first drafts. If you learn something too, then I can feel satisfied.

One last comment - the cover rocks. Thanks, Paul!

Unfortunately Mr. McShaffry fell into the same pitfall most other "here's THE right way to write code" authors fall into: he just has too many opinions and not enough facts. He touches on a lot of subjects which are programming religions, and there is no objective right or wrong for much of what he discusses. I feel some of the problem lies in the fact that McShaffry has worked on two types of games: the Ultima series and a playing card game. If you work on one codebase for years, you're going to think your solutions are perfect. As a person who has worked on numerous different types of games and engines I can tell you that there is no magic solution for writing the perfect engine. The engine is always very heavily tailored to the game desired. This is why you hear about the Quake engine, or some other third-party technology, being licensed and then gutted with major portions rewritten. So most of McShaffry's game-specific ideas need to be taken with a grain of salt. An example: smart pointers sure are safe, but if you're working on the PS2 you probably can't spare the memory or execution time for all that tracking.

I also disagree with his opening statements about variable and function naming. This is a constant headache for development teams, particularly game dev teams whose programmers are mostly self-taught. But just saying "Hungarian is useless!" is ridiculous. Sure Visual Studio tells you the type of variable, but this works maybe half the time (it is incredibly easy to confuse Microsoft's Intellitype), and not everyone uses VS. If a person is more comfortable in Emacs, should they be forced to change editors just to get context information on a variable? Is one to three letters of information in front of a variable name really that hard to type?

Those are just a couple of examples for the sake of brevity. My main objection about this book is that I don't feel I learned anything by reading it. I just read Mr. McShaffry's opinions about a few things, some of which had merit, but most of which weren't supported by any factual or reasonable data. They are just opinions. I did agree with his statement that all objects should have a stream-constructor defined; having tried to graft in "save anywhere" to two engines I can attest to the usefulness of this type of planning ahead. But other than that the book is mostly useless to me.

People new to the industry or curious about it might find it exciting to read tales from the trenches. And if you're a pro, I would expect your enthusiasm for the book to be directly relative to how closely your opinions match Mr. McShaffry's. But frankly I wouldn't recommend most of his ideas over anyone else's. Maybe it's the title that grates on me - perhaps a better title would have been "Game Coding: One Man's Opinion"

I've bought many game programming books over the years, and two authors stand out... Mike McShaffry and Andre LaMothe. This book is incredibly valuable as a reference and as a guide. Quite honestly, I wonder who paid the guy who wrote the "Spotlight Review" to dis it so badly, or who he paid to get his opinion in the spotlight.

But here's a test you can take for yourself... go to http://www.mcshaffry.com/GameCode and see how Mike McShaffry is *still* helping folks who've read his book, (or anyone who post on the site for that matter). He's still giving *free* advice on his book's forum, when most other authors won't even respond to an email.

In response to those who objected to the author's "coding opinions":
Yes, the guy has an opinon - he's entitled to. What do you expect from a book? "well, this is probably wrong, and I don't really know what I'm talking about, but the publisher paid me a lot of money so I have to say something." Give me a freaking break! OF COURSE the book is full of opinions - that's what books are!

Just one caveat - it doesn't teach you C++. It assumes some experience, meaning you can take the coding advice and apply it to suit your own style. It does assume a basic level of professional ability in other words.