Home Consulting Games Writing Personal Links

Fantasy War Designer's Notes

Most online games are poorly designed for online play.

That may sound like a contradiction--but it's true.

Most of the online games out there take one of two approaches. Many are basically solo PC games, modified for multiple players. Solo PC games are, of course, what computer game designers are most familiar with, and therefore what they look to first when imagining an online game. The problem with this is that online is a very different medium. In a solo PC game, you have a fast processor solely devoted to running the player's game; you mave microsecond access to gigabytes of data; your latency issues are minimal. In an online game, you have a server somewhere that can be overloaded if too many people show up to play; and you have the Internet, where latencies are normally in the range of a fraction of a second-- and where the speed of the modem limits the amount of data you can exchange in a short period of time.

What's amazing about games like Quake isn't that they're fun to play--what's amazing is that you can get a game like this to work over the Internet at all. An enormous amount of programing talent and effort has been spent trying to shoehorn a game style that works really, really well on a solo PC--or over an office LAN, which has much lower latency than the public Internet--onto the public net.

Even many online-only games perform the same technological hat-trick; Air Warrior, for instance, is a flight sim turned multiplayer. It works. It's cool. But again--it's a game struggling against the limitations of the Internet, rather than working with its strengths.

The second common approach? To take game styles that have been successful in ASCII, as free games, and pump them up with the multimedia assets that make so many CD-ROM games compelling--animation and video and music and sound. Ultima Online is the example: It's a graphical MUD. The problem with this is that MUDs work on the Internet only because they are ASCII; they require continuous connections to the client, which is resource intensive, but since only small amounts of data get exchanged, the resource hit is limited. Pump up the media, and you get a nightmare, a product that requires enormous hardware resources to support. And that means enormous costs. Given the fact that there's no working business model for online games, this is insanity.

I've designed games to be played with cardboard pieces and physical maps. I've designed games to be played with polyhedral dice, paper, and pencils. I've designed games to be played on solo PCs. And I've designed games to run on wonky, poorly designed proprietary online services (Prodigy, in this case). So I'm well aware that every medium has its own requirements. And that the smartest way to design a game is to look at the medium's requirements, identify its strengths--and design to take advantage of its strengths rather than struggle with its weaknesses.

As a simple illustration: You don't include complex mathematical algorithms in a boardgame. Why make your players haul out calculators? And you try to include continuous motion in a solo PC game; our culture is fixated on moving images, and the PC does that well.

So: What is the unique strength of the Internet as a medium? And how can we design a game that plays to its strengths--and avoids its weaknesses?

The strength is obvious: The Internet promotes communication. The thing people use the Internet for most is email. The web comes second. Chat and conferencing systems follow. A good online game should promote-- perhaps even require--communication among its players.

And indeed--that's an exciting prospect, because games are and ought to be social tools. Before the computer revolution, solo games were viewed as inherently inferior to "real" games--something to kill some time when you didn't have someone to play with. You'd rather play Bridge or Poker than Solitaire, right? The history of computer gaming has been strangely warped by the inherently single-user nature of the home PC, and the Internet allows gaming to return to its natural form.

How to promote communication? There are many ways to do that, and many potential online gaming styles. But one simple way is to design a game that depends on negotiation.

As a boardgame designer, I know how to do that, surely. I began playing Diplomacy as a teenager. And I've designed several diplomatic boardgames--Pax Britannica and Barbarian Kings among them. The key is to ensure that players are roughly equal in power, so that none can easily overcome an opponent without allies; to provide in-game ways for players to assist each other; and to ensure that alliances are never set in stone, so that players must continue to negotiate with each other throughout the game.

How do we allow negotiation over the Internet? Easy enough--chat in fast-paced games, email in slower ones. But that has another implication: the game must not be too fast paced. You can't negotiate in a Quake deathmatch, for instance; try it, and you'll get slagged while you're typing in a message to someone.

And how do we avoid the weaknesses of the Internet--bandwidth and latency? That's easy, in fact: We don't treat the medium as identical to a home PC, like everyone else. We treat it as its own thing. We design a game where bandwidth and latency just aren't issues.

Is that possible? It's trivial! Provide all necessary graphics and sound with the client. Design the game so it's turn-based, rather than real-time, so that clients only need to swap small amounts of data at infrequent intervals. Design it so that the server doesn't have to keep continuous connections open to every client, to minimize hits on the server and bandwidth concerns. Design it so turns update every few minutes--or even less frequently--so latency of a second or two is irrelevant to the player's experience.

Can you design a compelling game this way? Why not? What's the problem? I can design a compelling game with two paperclips and a rubber band. Tell me the constraints of the medium, and I'll give you the game. The trick is simply to stop trying to shoehorn everything into the genres you're familiar with. Everything does not have to be a shooter or a real-time strategy game.

I have no doubt that some people will have a hard time getting their head around Fantasy War. It doesn't have so many of the things people expect from computer games. There are no cut scenes here. Scant animation. Little sound. No continuous motion. No real-time 3D. No major advances in graphics. It does not push the technological envelope.

Try to write up a features list or a set of marketing hooks, and most PC game publishes would never fund this game.

But Fantasy War has, I hope, compelling attractions of its own, even if they aren't those that the computer games industry is used to. You can play it when you want, on your own schedule: a few minutes a night, or a couple of hours at a time. You get to make friends and communicate with new people. You have enough time to actually think about you strategy. You play in a colorful, richly detailed fantasy world. Your playing pieces have individual strengths and weaknesses.

There's nothing like this out there--more's the pity.

No one's figured out how to make money doing online games, yet. So why not take a risk with something different? The same old same old isn't working...

Sony is much to be commended for taking the risk.

And I hope you enjoy the results.


Nothing springs from whole cloth; and Fantasy War is no exception. An honest designer should acknowledge his sources.

Fantasy War owes a debt to a boardgame I designed almost twenty years ago, Barbarian Kings, long out of print. It, too, had individual heroes and wizards leading enormous armies to conquer the world; its play, though, was very different, strategy revolving mainly around the strict counter limit. And the world was no where near as rich as Fantasy War's.

We also owe a debt to Sylvan Clebsch's Stellar Crisis, a freeware game--in particular, the idea of "blitz" and "epic" games derives from it. It's remarkable that even with the enormous number of new online games in recent years, this wonky little freeware product remains one of the best on the net.

Of course any diplomatic game owes a debt to Allan Calhammer's Diplomacy, the granddaddy of diplomatic games.

And, on a personal note, I'd like to thank Dani Bunten Berry, Jessica Mulligan, and Brian Moriarty, who have done much to inform my own understanding of what's possible in this medium.

-- Greg Costikyan
New York
November 15th, 1998

Return to Games page.

Home Consulting Games Writing Personal Links

Copyright © 1998 by Sony Online Entertainment, Inc.