The Air-Speed Velocity of Unladen Gaming

Good Game Design

by Brad Wardell

Forget the marketing guys or the executive suits everyone claims are sucking the life out of the games industry. Sometimes the bloody trail left behind by a bad game that's DOA when it hits the retail shelves leads right back to the game designer. That's because designing a good game is tough, really tough, and what is fun on paper is too often tedious and boring in practice. The best games tend to be relatively simple, but game designers can't resist making complicated games. We're a bit crazy that way.

As a game designer myself, this has bitten me on a number of occasions. When we originally designed Entrepreneur, it had a ton more features than what actually made it out the door. Thatís because when we play tested those features, the game..well the game sucked. Let me explain: Entrepreneur was a real time business strategy game in which you start a company and must take over the worldís markets. You pick an industry to be in (like the computer industry or car industry or whatever) and then try to become a monopoly in it either by out selling your opponents or running your opponents out of business.

Well, in the original drafts, you could have as many products as you wanted and set the price of your product depending on where you were selling it. We also had a stock market in which you would try to buy stock in suppliers and other companies (like your competitors) so that you could potentially do a hostile take over. Moreover, you could research and patent technologies that others could use and have to pay for. Plus there was industrial espionage, employee relations to avoid strikes, defect management and more. Sounds great on paper if youíre a business sim guy but when we got it into beta, well, it wasnít fun at all, particularly if you wanted it to be a multiplayer game. What ended up happening is in the final months of development we jumped in and did emergency surgery on the game and replaced a lot of the most complicated things with what are called "Direct Action Cards" (think of Magic the Gathering cards for business). We got lucky in that the game turned out well. And next month when we release the sequel, The Corporate Machine, we were able to take what we learned and apply it to the sequel.

And this sort of thing happens all the time. The original design of Starcraft was far far more complicated than what was actually released as well. The Zerg, I believe, originally had a multi-stage process for handling their resources for instance.

Another example is again here at Stardock, we had been working on a design for a game to be called "Political Campaigner 2000". Basically, you were the campaign manager for a presidential candidate and had to help get them elected. It used data from polls in each state to determine how well you would do. We had to hold back on the game because we knew we wouldnít get it done by this election period. However, if we had simplified the design, we would have been able to get it out for this yearís election. Believe me, given whatís happened with the election, thereís some discontent that we didnít do that now. J

The bottom line is, many game designers stumble on wanting to put too many game elements into a game and forgetting that simplicity is often the key. It can cost lots of companies a lot of money.

So how do you make a good game? If you make it too simple, it wonít be "big" enough to be a major game. Itís a fine line between a fresh simple concept that you find at the store shelves and a $10 shareware game.

There are 5 simple rules to follow to make a successful (fun) game:

  1. Multiple paths to victory. Whether itís a strategy game or a platform game, you want players to be able to have multiple ways to win. In Civilization, you could win by conquering the world or sending your people to another planet. In Baldurís Gate II, you can play as both good and evil and most puzzles had multiple ways to complete it.

  2. Provide new things to discover over the course of the game. Donít put all your game elements up front. Have things you strive to get or see that makes it worthwhile to keep playing. Strategy games tend to do this by having new technologies and new units you can build. Role playing games keep you looking for that +5 armor.

  3. Keep the interface simple. Seems obvious but many game developers are more into the technology than the game. A good gameís interface shouldnít even be noticeable. If the player is losing due to not being able to navigate the interface, that will harm the game. I loved Total Annihilation but loathed TA: Kingdoms because of how much work it was to micro manage the units that you had to do to succeed.

  4. Avoid helpless defeat scenarios. You donít want players losing the game because of something they consider cheesy. In Starcraft, it could be frustrating having a mass of cloaked ships wipe someone out because their otherwise impressive army didnít have enough cloak detection units. Total Annihilation had the Big Bertha which could be abused. Similarly, in role playing games, you have to keep players from getting too far too fast to where they canít defeat the bad guys. Believe me, this is not a trivial thing to design in. Cheese tactics are very hard to design against but the games that stay popular over time are the ones that successfully design against it.

  5. Design for the proper target hardware platform. Yea, I might have a dual 850 setup but I suspect most people donít. We design games for P2-233 systems. Unless youíre making a totally cutting edge game (like a first person shooter) you better make sure that most people can play it optimally. Strategy games for instance, arenít 3D because people want them, theyíre that way because game developers want to make 3D stuff. Thereís no excuse to double the hardware requirements for your strategy or RPG to make it a 3D engine. There is market research out there and believe me, people donít care whether the engine is 3D or 2D when they make purchasing decisions.

It is easy of course to sit back and talk about what people should do. Hey, I know these rules and have blown it before myself. And it gets harder each year as the pressure to have multiplayer gaming in everything grows. Allowing an air transport unit to pick up an opponentís unit (like in TA) wouldnít be a big deal if it were single player only. But in a world filled with insecure zitty 15 year olds taking out their aggressions on the world, such minor issues have to be patched.

As with most people in the game industry, I regularly get letters and email from users making suggestions for what we should do in a sequel to our games or submitting their own game designs. And it is pretty universal that people want to put a heck of a lot more features in the game than would be good for them.

In fact, the very features we had taken out of Entrepreneur are regularly requested as "new features" we should put into The Corporate Machine. Next issue weíll talk about the perfect real time strategy gameÖ

Brad Wardell is the Project Manager of two ongoing Stardock games:

The Corporate Machine, a business strategy game due out December 11 and Galactic Civilizations, a turn based strategy game due out late 2001. Stardockís website is http://www.stardock.com. Brad Wardellís website is http://people.mw.mediaone.net/bwardell. He is also the Product Manager of WindowBlinds (www.windowblinds.net) and DesktopX (www.desktopx.net).