Playing Dark Souls has gotten me thinking about the different ways that games structure their world. I really want to describe, in a concrete way, what makes exploring the world of Dark Souls feel like such a different experience from most game I’ve played, what made it feel like such a unique space.
The first thing I thought about was freedom. There’s undeniably a prevailing trend in game design to force the player down fairly linear paths, where he works his way through one environment, then another, and so forth, until the end of the game is reached. Dark Souls is open to the player to approach in any order, as long as the obstacles blocking the way are cleared or bypassed. It gives the player a large degree of freedom to choose how he wants to approach the game. It’s hardly unique in this regard: Sandbox titles like GTA afford the player this freedom, as do Metroid/Castlevania style exploration platformers. Dark Souls is certainly similar to those games in some ways, but also distinctly different somehow…
So, next, I focused on continuity. Most areas in games are broken into discrete chunks, connected to each other by loading zones which may or may not represent an actual spatial relationship. It’s often not clear where these environments are in relation to each other, nor is it usually considered important. Even games which give the players a great deal of freedom, such as Deus Ex or Metroid, still have at-times arbitrary level breaks which separate areas from each other and give the player no real idea of the space being covered. Dark Souls takes place all in one large area: Beautiful backdrops which, in most games, would be mere pleasant looking skyboxes, end up being the next challenge area after a few hours of gameplay. This is another aspect in which sandboxes like GTA offer something very similar– and yet, exploring the world of Dark Souls still, somehow, feels completely different…
This gives three metrics to evaluate way games use their world space:
Freedom vs Guidance
Continuity vs Segmentation
Complexity vs Accessibility
Let’s look at some games from the perspective afforded by these concepts.
Dark Souls has a high degree of freedom, but there are certainly games which offer more, since the player is restricted in a number of ways based on what routes are opened up. However, all restraints on the player’s movement are within the diegesis of the game, so it rarely feels unnaturally constricting. Dark Souls is nearly perfectly continuous, with each part of the world having a spatial relationship to each other part– there are a couple of exceptions to this but these, too, are within the diegesis, special portals which lead to mystical worlds that would be inaccessible by any other method. Dark Souls is also highly complex, the continuous world is twisted in upon itself and requires complicated detours to navigate. All of these together produce a very unusual and rewarding environment to explore, but one that can easily become overwhelming and confusing.
Half-Life 1 and 2 have moderate levels of freedom: There is always a pre-determined path forwards, but each area has room for exploration and is amenable to several different approaches. The second game has a lower level of freedom in general, but the two are comparable. They have middling degrees of segmentation, with the second a bit more questionable in this regard: Whereas in the first game the environments, at least early on, have a definite and understandable spatial relationship with each other, in the second game each area is so strung out and linear that we kind of have to take the developer’s word that these places are spatially related. Finally, they are of moderate complexity, with navigating tricky environments being one of the primary challenges faced in both games– although, because of its more segmented design (and its infatuation with its at-the-time outstanding physics engine), the environmental challenges in the second game tend to come off as a bit more contrived. The overall effect is very much as though someone has outlined an adventure and you’re just filling in the detail
The Grand Theft Auto games are an interesting example, since they provide both a low-moderate freedom main story path and an extremely high freedom sandbox which both occupy the same space. The world is perfectly continuous, with every game location within the boundaries of the game world having a clear spatial relationship to every other location, though the boundaries themselves seem at times to be rather artificial. These game worlds are also extremely accessible, flat and easy to navigate: For most locations within the game world, once you know where they are it is a trivial challenge to reach them. Navigating the environment is easy and can be done in a number of ways.
The split approach to freedom/guidance in the GTA games provides an experience that is sometimes schizophrenic, but also allows different kinds of players to take what the game has to offer at their own pace.
I think those examples are particularly interesting, but one could view any spatial game in this manner and perhaps gain some valuable insight into why these environments are interesting to exist in (or not, as the case may be). Heck, maybe there’s a whole extra axis one could analyze environments on that would be just as interesting, but one that simply didn’t occur to me.
I don’t make any value judgments here, and I think that freedom is as interesting as guidance, complexity is as useful as accessibility, and continuity offers different benefits than segmentation. However, that being said, I certainly believe that, right now, most games out tend to overrepresent and overprioritize guidance, accessibility, and segmentation. It’s worth taking a long hard look at the benefits that can be reaped from both ends of each of these three spectra, and to design your game world accordingly.