There's a fun thread raging in the GameSpy Forums about Great Game Ideas that Just Didn't Cut the Mustard. It started in July and people are still actively tossing opinions into it. What's fun is that one man's masterpiece is another man's failure, depending on what you wanted to get out of the game. It's a great read. I love the bitterness:

"RAINBOW F---ING SIX! You know, instead of planning out the most strategically advantageous route for your team to follow, you end up planning out the best route that your retarded AI teammates will take without f---ing everything up." -Pfhor-Chan

It's tough to be a game designer these days. Twenty-five years ago, if you had a crazy novel new game idea, it took about a month of work to fill up the 64K of system memory computers had available, then you could shove your idea into a zip-lock baggie and start selling it. Who cares if the idea didn't work? Hell, after tinkering with it for a couple of weeks, you could bail on the idea if it wasn't coming together and who cared? Nobody.

Nowadays, game development is like trying to steer the Titanic. You come up with a game design ("Hey guys, let's try sailing to New York!") and then you try to make it happen. Building a modern triple-a game takes about 50 people, contributing about 2-3 years. Budgets are in the tens of millions of dollars. What if the design just isn't working? It's iceberg time. That's why publishers are eager to greenlight games that are clones of existing games -- hey, at least you know it'll be fun.

Peter Molyneux, for better or for worse, is always steaming full speed ahead with inventive game ideas, even if they may or may not ultimately work out. Fable 2 is nowhere near finished, according to Kotaku, and it's already a couple of years into development. Will it dazzle? I'd be freaked out if I had invested millions into the development, but since I'm just Joe Gamer, I can't wait to see what comes out of it.

Game developers are wrestling with how to still be innovative given how complex it is to build modern games. Frequent and rapid prototyping is one answer. That's how Will Wright's team is building Spore. There's also a great lecture on prototyping by indie game experimental artist Jon Blow, which is interesting to watch if you're into the indie gaming scene. (That's where all the innovative game ideas are these days... although Nintendo deserves some credit for shaking game designs up with the Wii controller.)

Along those lines, there's a development process called 'SCRUM,' which is basically a software design philosophy that's been rippling through the game development community for a while now. Gamasutra just posted a piece explaining SCRUM. It works like this: Instead of defining a goal and then reaching it in a series of milestones, you begin with a rough outline of what you want to build, get a prototype up quickly, and then frequently review and reevaluate the product throughout the process. Your final design might be very different than what you set out to build, but it'll probably be a much better product. Whether or not this approach actually results in great games ... well, that hasn't really been proven yet. And it's hard to implement Scrum across a whole developer and publisher and have everyone agree on how it should work. Here's a fun party trick: if you ever meet a game developer at a party, say the word "Scrum" and see if he or she bursts into tears.

If you're like most gamers, you don't care how a game is made ... so long as it's cool. You've probably noticed that certain developers -- like Valve or Blizzard -- consistently come out with games that are beyond awesome. Why is that? Is there some secret mojo stored in vats below their building? Personally, I think what separates these developers from the rest is a willingness to outright terminate projects that just aren't making the grade. Look at Valve Software: TeamFortress 2 has been cancelled and resurrected at least once, and the result looks spectacular. Blizzard, meanwhile, has cancelled games (WarCraft Adventures), delayed them (StarCraft Ghost), or rebuilt them from scratch (The original StarCraft was scrapped and reinvented before release.) There's an article in the Orange County Register about it. Of course, cancelling projects requires both brass balls and deep pockets, so developers with the resources to do it are few and far between.

Keep that in mind as you enjoy the GameSpy thread and reminisce about game designs that were almost great.

      -Fargo