The Rise of the Systemic Game | Game Maker’s Toolkit


When it rains in The Legend of Zelda: Breath
of the Wild, everything changes. Rocks become slick, which makes them difficult
to climb. But Link’s footsteps are hushed, making
it easier to sneak about. It dramatically boosts the damage of electrical
attacks, but it turns fire and bomb arrows into regular arrows. Certain plants and creatures, like the electric
darner, only appear when it’s raining. Human characters run for nearby cover. Campfires sizzle out. And you’ll even find massive great puddles
in certain areas – which then evaporate when the sun comes out. The rolling weather system in Breath of the
Wild is not just a nifty visual trick, but something that can reach out and influence
almost everything else in the world. And this is the definition of a systemic game. As Aleissia Laidacker, a former lead programmer
at Ubisoft puts it: ALEISSIA LAIDACKER: Systemic means there’s a link between all the systems in your game. They’ve been developed and designed with
the intention that one can influence the other. And in the last few years we’ve seen an
explosion of games that have this sort of interconnectivity – from Japanese titles like
Metal Gear Solid V and Zelda, to ambitious European games like Kingdom Come: Deliverance,
to smaller indie titles like Mark of the Ninja, to pretty much everything Ubisoft’s making
at the moment. Take Far Cry, for example, where much of the
joy comes from the fact that not only do you fight enemies, and not only do you fight wild
animals – but enemies can fight animals. And vice versa. This is not a scripted encounter, hand crafted
by the developer. It’s just the enemy system and the wildlife
system interacting… as a tiger mauls a guard’s face clean off. This works thanks to “awareness” and “rules”. In super simplified terms, every entity in the
game has inputs, which are things it can “listen” for. In the case of the Far Cry tiger, that might
be the player, bait, fire, and enemies. And entities also have outputs, which is when
they broadcast their existence out into the world. If the input and the output match, and the
entities can see or touch each other, a connection is made. And that’s when a rule is followed. In the case of the tiger and the enemy, that
rule is face mauling. But for other games, it could be that if a
flame touches a wooden arrow, it sets on fire. Or if an orange tree frog explodes near a
floor tile, it’s destroyed. Or if it rains on a camp fire, it sizzles
out. So systemic games work by having all sorts
of objects, characters, bits of the environment, and game systems be aware of each other, and
have rules for how to interact. But why should we care? What makes this style of design more interesting
than games where the systems aren’t so connected? Well, one huge advantage is that the player
can make interesting plans. In a lot of traditional games, entities are
only really aware of the player and not much else. Meaning the only way to interact with an enemy
is through very direct means. A.K.A shooting them. But because entities in systemic games are
aware of so many more things, we can get to enemies through indirect means. Like, releasing a caged animal to have it
attack nearby guards. So a defining trait of systemic games is that
you can exploit those relationships between different entities and systems, as part of
plans – which make you feel rather smart when they come together. Here’s a good one from Watch Dogs 2. I was getting chased by the cops, so I lead
the police into gang territory, hid on a rooftop and watched those opposing factions fight
it out, and then used the distraction to high tail it out of there and lose my wanted level. That felt great, and a lot more interesting
than just shooting a bunch of cops. Another massive advantage of systemic design
is that these games can create moments of drama and surprise. Such as a wild three-way tussle between the
royal army, the golden path, and an angry elephant Again – this wasn’t scripted, but something
that occurred organically from a bunch of different entities that are all aware of each
other, have rules for dealing with each other, and have found themselves in the same place. And I think these anecdotes are cool for two
reasons. For one, they often follow a really interesting
story structure as you devise a plan and try to execute it, but a surprising chain reaction
of events scuppers your plan and you must react and adapt. And because these anecdotes are completely
unique to your experience, I think that often makes them more special and memorable than
the ultra epic moment that every single player is going to see happen. No one’s tweeting about that, are they? So systemic games allow the player to come
up with interesting plans. And they lead to surprising anecdotes. And we call this stuff “emergent gameplay”
– things that weren’t intentionally designed by the game’s makers, but solutions and
situations that emerge thanks to the meeting of multiple systems. That doesn’t let game makers off the hook,
though. They’ve got to set all this stuff up, to
create the possibility of emergence. So to make this work, we need to create awareness
between lots of different entities in the game. The more things that are aware of each other,
the better. Having characters be able to fight amongst
themselves is one thing, but we can also have enemies be able to damage the environment,
have entities be aware of systems like the day and night cycle, or – in the case of Zelda
– invent a whole chemistry engine with wind, fire, ice, and so on. Next, we need rules that are consistent. Because – as I talked about in the AI episode
– you can only make good plans if you have a pretty strong idea of how the system will
react when you push on it. But that means we also need rules that are
universal. Like, if some wooden objects catch on fire,
then all wooden objects should catch on fire. Any time a rule like that is broken, the believability
of the world is reduced and the player is bit less likely to experiment in future. So to make this sort of universal connectivity
easier, perhaps take this advice from Dishonored’s co-directors Harvey Smith and Raphael Colantonio. They say that instead of having entities be
able to listen for specific objects and characters in the world, they let entities output general
stimuli – like fire damage, piercing damage, or explosive damage – and then have other
things in the game be aware of those. This layer of abstraction makes it easier
to add and modify entities, and even allows players to find connections that the designers
never even thought of. Like in Harvey Smith’s most famous game,
Deus Ex, where you can exploit the fact that MIB enemies explode upon death to blow your
way through doors. This is also why you can lay metal objects
on the floor in Zelda to conduct electricity and solve some of the shrines in your own
way. It might feel a bit like cheating, but that’s
what makes systemic games fun. Instead of finding the single, authored solution
to a puzzle, you can use the inherent behaviours of the game’s systems to find your own way
to overcome the problem at hand. And then, if you want surprising events to
happen on the regular, it’s important for the system to be somewhat unstable. To not be in perfect equilibrium until the
player comes along and pushes on it, but capable of moving and shifting all by itself. This can be achieved with automated systems
like how the rolling weather in Breath of the Wild can create unpredictable thunderstorms. Or by giving AI their own goals and needs
which will push them to move about and, hopefully, into conflict with other entities. Okay. So you’ve made a systemic game. You’ve created connections between all the
entities in your game, and they follow consistent and universal rules of interaction. But there’s still work to do, because it’s
easy to screw these sorts of games up. Many fail to truly encourage players to find
these emergent solutions. Because, despite all of the exciting opportunities
afforded by a game like Metal Gear Solid V, I ended up finishing a lot of missions by
abusing the silenced tranquilliser gun. Why risk some ridiculous plan if there’s
a much more reliable solution to the game’s challenges? Hitman pushes you to be imaginative by making
Agent 47 weak in a straight up firefight. And Zelda, controversially, makes your weapon
turn to dust in the hopes of pushing you to try more creative solutions. It’s also important to give the player a
whole range of ways to interact with the world, that go beyond just killing everything. Killing enemies essentially removes an entity
from the space, which most often reduces the possibility for systemic fun. You want to give the player tools that let
them change, or even add entities – not just remove them. That might mean hacking a security system
to change its allegiance, or creating an inflatable decoy to distract enemies. Other games screw up by constraining the player’s
options in really linear missions. Grand Theft Auto is a right pain for this. Those games are filled with good systems,
to ensure that the cities feel alive and realistic. And the police wanted level is an absolute
stroke of genius. Without it, killing a civilian in GTA would
have no interesting impact on the world. But because your murder feeds into this wanted
system, which sends cop cars after you, your action actually has consequences that ripple
out into the different systems in the game. It’s terrific. But the game’s main missions are often incredibly
linear with scripted sequences and have all sorts of fail states for not perfectly following
the commands on screen. Far Cry has this problem too, where the main
missions are often nowhere near as much fun as the camps – which are just open-ended testbeds
for the different systems. Making levels for systemic games is more about
giving the player a goal, and not caring how they achieve it. They require open areas, and lots of entities
that are aware of each other to create the opportunities for good plans and memorable
anecdotes. Also, some systemic games fail to create a
unique experience. Ubisoft, bless their hearts, are trying to
fill all of their games with the ingredients for emergent thrills but the camps in Assassin’s
Creed Origins feel remarkably similar to those in Far Cry 4. So it’s important to set the systems up
so they deliver their own experience. Look at the rather different Far Cry 2, where
it feels like the systems – such as fire propagation and roving bad guys – are designed primarily
to whip up moments of peril and danger for the player. Whereas assassination simulator Hitman goes
completely in the other direction, with the focus being more on having a perfectly choreographed
system, where the player becomes the spanner in the works. You can even imbue systems with a message,
like in Mafia 3 where the police react to crime less quickly in a black neighbourhood
than a white neighbourhood. That’s a game speaking through its systems. Now, this sort of systemic design is really nothing
new. For decades, simulation games have been using
this sort of interconnectivity to mimic real world systems. They just happened to be games where you play
as some sort of omnipotent being, looking down on things. And we’re still seeing this sort of stuff
today with games like Rimworld, and the absurdly interconnected Dwarf Fortress. In that game, you can have an emergent situation
where cats wind up dying – because dwarves splash wine onto them while drinking, the
cats drink the wine while cleaning themselves, and then end up dying of alcohol poisoning. Ridiculous. Watch this episode of Eurogamer’s Here’s
A Thing for more on that. But then there was the immersive sim – the
genre of games like Thief and Deus Ex – where the whole point was that they took that simulation
design and put it into an immersive, first-person game where you control a single character. Hence, the name. And this is a big reason why I was stoked
for the comeback of the immersive sim – with great games like Prey and Dishonored 2 – but
pretty bummed out when it seemed like those games weren’t connecting with a lot of people. Sales wise. But now it’s starting to become clear that
this sort of design is popular, it just doesn’t have to be limited to that very specific legacy
of games that stretches back to Ultima Underworld and System Shock. Whether it’s big budget experiences. Or janky European games. Or indie titles, made by immersive sim fans. Or – absolutely weirdest of all – the latest
Legend of Zelda game, we’re seeing systemic design and emergent gameplay rise up and appear
in all sorts of games. And as someone who loves this sort of stuff
– for the opportunity to make plans, and then see them go horribly wrong – I’m very excited
to see where this trend goes next. Thank you for watching! Systemic game design is a super complicated
topic so I’m only scratching the surface in this video. But check out the description below for links
to loads of resources from experts in the field.


Leave a Reply

Your email address will not be published. Required fields are marked *