Wednesday, June 09, 2010
Perfect MMORPG: Design fundamentals
In the open Sunday thread there was a discussion whether a developer should design a MMORPG by following his vision, or whether he should make what the players want. Well, I have a rather healthy distrust of both of these methods. When I think of developers with a vision, I think of Paul Barnett and the famous Bears, Bears, Bears story: I think the developers being very public about their great vision for Warhammer Online, and the actual game not living up to that vision, has hurt WAR far more than it helped.
Doing what the players want is downright impossible. What they players *say* they want, what they *really* enjoy, and what they end up doing in an actual game are three very different things. Half of what everybody thinks that players want is simply not true, and based on observation of players who by various in-game rewards got persuaded to act against their own best interest. For example a great number of people, including professional developers, think that "players want to solo". I'm pretty sure that players enjoy the *option* to solo, and a few lone wolfs solo all the time. But given half a chance, most players would group a lot more, if the incentive structure and logistics for that work out. Just look at World of Warcraft pre-Dungeon Finder and post-Dungeon Finder: The number of people spending time in a group has increased dramatically, both while leveling and at the level cap.
So I think the way a game should be designed is from the bottom up, with solid engineering instead of nebulous visions and misconceptions about what players want. A developer needs to be able to answer questions like: The game has launched 6 months ago, it is 8 pm, prime time, there are 3,000 players on the server; where exactly are these players, why are they there, and what exactly are they doing?
To answer such questions a developer needs to "map" his game. Instead of having just a list of features, there needs to be a diagram showing how these features relate to each other, and what the flow of gameplay through these features looks like. A MMORPG has "basic repeating units", like combat, which are the core of the game, and which most of players will spend most of their time doing. These basic repeating units are held together and motivated by the next layer, e.g. quests that tell you to kill 10 monsters and give you some reward for that. And in the outermost layer there is the virtual world with its zones, quest hub locations, dungeons, and lore.
Every possible feature has to fit into that map. You can't just take a developers "vision" of "I want player housing", or follow some players' demand for such a feature. You need to be able to answer questions like what players are supposed to actually do in their houses, whether there is a basic repeating unit of gameplay like crafting or redecorating involved, and whether lots of players sitting in instanced houses will not make the open world look deserted.
And while in my examples I used descriptions of that map which correspond to many current MMORPGs including World of Warcraft, I do think that to make a really successful game, the gameplay map of it has to look significantly different from World of Warcraft and similar games. That starts with the most used basic repeating unit, combat, having to be noticeably different from "target mob, hit the same sequence of hotkeys over and over". On the next layer there should be motivation different from "kill 10 foozles" quests, and so on.
I do believe that the success of World of Warcraft is due to the solid engineering of their gameplay map. But the answer how to make an equally successful game is not to do reverse engineering and then using a copy of that same gameplay map for your own game. It has to be designing a *new* gameplay map, from the ground up.