@SamTung shared the CEDEC 2017 Zelda talk (the video is paywalled so we have to make do with photos) in a Slack convo and I thought it’d be worthwhile to discuss on here.
To start out with, I just want to state the obvious point that Breath of the Wild is an OSR take on Zelda. Unlike rogue-likes it doesn’t use a hex-crawl style procedurally generated world (see for example Caves of Qud), but it does resembles hex crawls in the way its encounters are highly systems and procedurally driven, and the way it encourages open-ended player exploration of a “wild” space. It also leans on OSR concepts like player ingenuity in problem solving, resource scarcity and foraging, encumbrance, and equipment degradation (See Macchiato Monsters for an excellent implementation of a simple degradation mechanic).
Accordingly, OSR design can probably learn a thing or two from this mammoth multi-year design effort by one of the best design groups in the world.
The key design breakthrough that the Nintendo devs shared in their CEDEC talk was assigning a subjective “gravity” score to each object on the map. These gravity scores represent the degree to which the object pulls player interest. We can think of them as similar to The Order of Might in Torchbearer, HD in a lot of OSR games, or Tier in Blades in the Dark.
By balancing the gravity distribution across the (curated) map, the Zelda devs were able to effect a dramatic change in player behaviour, where exploration was far more evenly distributed.
So what can we learn from this stunning development? Here are some of my first thoughts:
-
Space and time in BotW are continuous, whereas in TRPGs they are more or less discreet. A dungeon crawl is the least discreet representation of time and space in TRPGs, a timelining game like Microscope is the most discreet, with a point crawl occupying a place in the middle (Time progresses in a linear but interrupted way, with space being totally abstracted between points of interest). This makes a big difference in design values.
-
Gravity is mostly a useful design concept in games concerned with a sense of world objectivity. Highly collaborative games won’t get much use out of it, because concerns about gravity are implicitly resolved in conversation dynamics. Reading other players and having a sense of established narrative tropes is much more valuable in these games than some numerical ranking of interest. Even if rankings are used, they tend to be one-dimensional for ease of use, unlike the two-dimensional system used in BotW (because it’s on a map).
-
Gravity is communicated visually in BotW (mostly by size), whereas it may be communicated verbally in TRPGs.
- For example, in a hex crawl a group ascending a hill may reveal 2 hexes out of the fog of war to represent an expanded visual horizon. The map may actually have some icons on it to represent objects and some gravity can be communicated that way, but the GM’s description of what is seen will always be more detailed than what is visually communicated.
- Grid based dungeon maps can communicate more in this regard, but they tend towards visual sparseness to maintain a sense of the unknown. The detailed pictures in dungeon crawls tend to be shown to the players once they’ve already reached a point of interest, so any consideration of gravity is moot.
- Some point crawls like Slumbering Ursine Dunes do use a rudimentary sense of gravity to visually mark major points of interest and draw the players in different directions across the map.
(follow link for image)
- Gravity might be best used on a “second pass” of making a bespoke map by hand. Your map might look visually balanced and conform to geological rules of thumb, but is it really gravitationally balanced? Try ranking your points of interest on a numerical scale and make sure their ratios of quantitative gravitational difference are roughly matched by their ratios of spatial difference (number of hexes apart, number of pixels apart on a point crawl).
- We already weight encounter tables by gravity (Again, Macchiato Monsters’ risk die implementation is excellent here), and this does help with making procedurally generated game spaces that have a satisfying variety and texture, they don’t do much to add to player foreknowledge and choice (encounters tend to be separate from one another).
- One useful bridge of encounter tables and spatial generation might be to combine the world generation approach of The Perilous Wilds with the weighting system of Macchiato Monsters. The Perilous Wilds abstracts space in a way similar to a point crawl, but its generation principles might be applied to various representations of space.
- I’m also interested in thinking of how these gravitational principles can be applied to games with less emphasis on traversal like Blades in the Dark and Urban Shadows. In these games the space of the city matters because the city as a space is narratively meaningful (unlike for example Monsterhearts or Apocalypse World where the setting matters, but its spatial existence can be hand waved away with no problems). Could we use gravity to help spatialize the distribution of events in the city in such a way as to help encourage a sense of the city as a whole instead of focusing in on a narrow part of it?
Anyhow there are some thoughts on the gravity concept, what do you think about it?
