Pursuing Interactive Suspension of Disbelief

oh don't clap yet' hello yeah my name's Shawn I work at a studio called Campo Santo hold on I have no presenter notes and you'll just all the standing here in silence for 25 minutes cool and before that I worked at telltale games and while I was there I was lucky enough to work on games like tells a monkey island poker night at the inventory personal favorite of mine Nelson tethers puzzle agent and thank you and the walking dead season 1 and yeah so I'm currently working on a game called fire watch with everyone at Campo Santo and it's a first-person adventure type of thing and the games core focus is the relationship between a man and a woman and I'm gonna it's in isolation sorry so I can talk about that more a little bit we're going to talk about some of the challenges you're facing but even

though all these games are very different experiences I kind of believe they're all after the same thing and they kind of focus on a few narrative concepts one is character interaction with realistic believable dialogue also stories and anecdotes the player can discover an impact in a meaningful way and interaction systems that pay attention to what the player is doing and the choices they're making and looking over all these games that I've been lucky enough to work on the most recent being The Walking Dead I see that these concepts are all working towards a common goal and the talk the suspension of disbelief there's been a sort of concerted effort on the part of these games to presuppose that they're not games or puzzles or whatever that there are simply experiences they're meant to entertain provoke and surprise and I think if you're trying to make a piece of fiction about people that the suspension of disbelief is really one of

the most important things to consider in your design so for a quick turn to Wikipedia the phrase is coin in 1817 by poet philosopher hypochondriac opium addict and cravat enthusiast samuel taylor coleridge who suggested that by being sure to anchor and otherwise impossible tale in elements of the human experience readers would suspend their judgment of the fantasy and experience it for its drama so Coleridge was working during the time of the Enlightenment when people had begun to stop believing and things like witches ghosts inspectors in their everyday lives so consequently these absurdities had fallen out of vogue and popular works the suspension of disbelief as a tool for the author became a way for these very powerful images to bridge the gap between what was popularly believed and the thematic meanings the images could

bring with them such as the spirits that Coleridge himself employs in his poem The Rime of the Ancient Mariner the audience could see past their disbelief of the supernatural if they're actually being told a story about people it was the author's job to assure that that was the case now we jump forward to the age of cinema and it seems as if the onus of the suspension of disbelief has sort of transferred to the audience so the audience to really engage the human elements of a story they are the ones that are asked to take the supernatural at face value to ignore the artifice and to suss out the meaningful parts of the story the truth I mean the same is often really said about video games just ignore certain aspects of the system or the gameplay or the dialogue with very premise itself and it's there you're gonna find enjoyment and meaning and value like I'm not saying you take this literally like I'm not suggesting that we make like hyper immersive like completely real simulations where the

audience actually believes this is happening and therefore they feel the feelings that are inside the text what I'm saying is that I'm with Coleridge I reject the notion that the player should have to ignore certain aspects of the experience just to find value in the narrative or to connect to the story and if you're gonna make games where the narrative matters and then I think it's the job of people in this room to communicate to the player through and focus on the human experience that the game the player is set down to is worth engaging with emotionally and not just take for granted that the player actually should so Jake rod can who's Campo Santo is co-founder and I have described video game design in this method as like going on a date with a player like the players at the table with you and for the date to go really well you have to listen to what that person is saying or what that person is expressing and then respond with a mix of comfortably predictable and like surprised excitingly surprising ways Gabe Newell has referred to valve sort

of design and marketing philosophy as a dance with the audience valve is gonna take the first step like they're gonna put the game out or design the IP or the system or the backend or whatever and it's the player that will then sort of find meaning in it contribute to it through play or oftentimes even there like an asset creation and then the company has to respond so this understanding that you're dancing with the player I think is key in creating an interactive piece that suspends the audience disbelief so I would quickly when I look at the Walking Dead and think about like how we sort of did with that and I think we set out to strip away things from the classic adventure game that forced me to consider its structure and we really wanted to replace that with character interactions that drove the entire experience forward we wanted the players to be a participant not an all-powerful entity because we felt that was appropriate for the property and was just sort of interesting to us as designers so when we consider these narrative concepts

character interactions dynamically discovered story and player where interaction systems we feel like the walk and get actually did pretty good we came at the game with a certain assumptions about adventure games that dialogue trees were static that story lines didn't change and said well what if they did and what if you don't answer and the game just keeps going on and also what if players have radically different outcomes and I feel like the game enabled most gamers to see past the artifice of the experience and just sort of consider it not as a game but as a human story experience even for if only for a few minutes here and there so where does the game fail firstly I think like I could have and like a bit of like that gets the writing in general could have structured the dialogue responses throughout the game to be more responsive whenever the game is obviously responding to a choice you're making it feels like it's really succeeding but there are times when either if you sort of just like scope or like personal laziness that the game

feels like it's just beating you core story content and then you can kind of see the players eyes glass over because they're experiencing a story but they're not experiencing their story and I also think the game falls down through the use of QuickTime events which we all unfortunately know seem to be really way too often employed in games with narrative focus but I do want to talk about those because when I was considering failures for The Walking Dead for this talk like he's obvious like the first thing I wrote down but then I think the reason that QuickTime events are universally maligned is if because of something inherent to their design I think it's because of their misuse so I'm going to like look at a couple examples so here hopefully this works yeah the player has been startled by a zombie during their first conversation with Clementine and it's like jarring it is orienting it's a situation where Lee will escape by the skin of his teeth but it's totally okay if you die you're

gonna get better at it it's a mix of mashing and grabbing and object use most of which are tied to the game's core mechanic the reticle so the player is gonna use that to go on to select basic objects shoot guns and open doors it's disorienting in a forgiving way and also teaches the player to become more proficient at the game's core mechanic pointing and clicking and then later did that work though maybe it'll work okay we subvert the QuickTime event in a fight with a stranger we drop the UI from the screen and we actually let the player continue to choke a man to death if they choose we also utilize Wickham events when you're supposed to fail trying to open a shuttered door or even just trying to stand as you're moments away from dying and there's like this emotionally and I think that's really emotionally effective we're doing our best given the system to the game to share with the player leaves ever increase lack of power however we also started using them namely dislike button presses and matches like really broadly press X to dodge and like excuse me press B to

matte like open a thing and those I found sort of game fi the experience and they don't serve to continue that dance with the player it's like stepping on the players foot like you both stumbled Mike literally look down and to see what's happening at the controller level and like we tried of course correct and I need to spend a bunch of time and mental energy to get back in time with the music and getting back in time of the tone of the story like essentially QTEs are not the problem it's like how when and why they are written into the game and how they impact the overall tonal consistency of the experience so I want to talk for a second about tunnel ins consistency it's incredibly important concept to the suspension of disbelief and I don't wanna go too far in the weeds on that I would point you towards like Clint Hawking's talks at GDC before Steve Danner does a podcast called tone control that I think is very good and if you want to go into that but like anything else you have to consider your writing and the narrative is a consistent piece of the whole and make sure that at every level it's working

towards the creation of a cohesive experience my argument is that sometimes QuickTime events actually do that but used broadly they won't the issue is usually the QTEs destroy the mechanical cohesiveness of the game in turn obliterating the players suspension of disbelief and undercutting the importance importance of the game's story and the players mind along with it so tonal consistency through theme mechanic is one of our chief concerns on the game of working on fire logic Campo Santo you again you plays a man named Henry who's retreated from a sort of like messy life to be a fire lookout for the summer and the absorber wilderness and Wyoming and while doing the job you build a relationship with your superior woman named Delilah whom you talk to exclusively over a handheld radio but on your first day like something weird happens and you have to go out of your tower and then into the when you go out into the world and through environmental exploration and sort of narrative choice you learn to inhabit Henry and build a relationship with Delilah we're super early so names and things might change but hopefully not because we keep saying

them but um we'll see but it's a first-person story experience and we're most games like this would put a voice in your ear and like expect you to follow it and we do we just thought it'd be interesting if you could talk back so in this game you can a large personal goal for the project is to suspend his suspenders belief obviously in order for folks to see sort of past the artifice of the game and feel like they're having a real conversation with Delilah so if players be like they're having real conversations and like that's great like gold number one checked off then the next one is the folk for folks to feel like they're building a relationship with her and if we can do those two things and hopefully we've created a game where the player feels totally in control of the action and they're pushing and pulling on the emotional aspects of the story in a world that is responsive and fulfilling like that's the entire goal of the project so we're gonna be talking to Sarah aside we're doing what the goals of a project we need is like always anxious about making it like it's I've never done this before it's strange but

uh to that end what are some of the things that can stop us from achieving that goal I have a lot for the written dialogue being unresponsive to player choice I'm just gonna go through them poor voice acting is a good one checkpoints missions arbitrary game elements such as these the list goes on and on poor density of world interactions is a big one unresponsive world state for president of dialog interface tutorials and UI failing to pursue empathetic character is it just the list of things will stop us from succeeding it seems infinite but some of these things are actually avoidable through disliked execution that's just like talented people are doing their job as well which is kind of hard to comment on but other things are solved through game design and then some of these things though are solve not just through writing a form of execution but how we write and when I talk about how we write I'm talking about sort of the principles we use when writing dialogue in addition to the tools we develop to organize and implement the

stuff that we're gonna make because one of the things like one of things I personally remember when I'm either like at any stage of development is that nobody really cares about writing like in the traditional sense like people just don't recognize objectively strong dialogue if it's presented today statically or like in a laborious cutscene or not really obviously linear way but if you make if you let the player make a choice to deliver a particularly well-written line then they're like we're often told the writing's really good so that line the exact same line talked into like an expertly crafted 90 second cutscene gets completely dismissed if it's ever seen at all because the player just feels like they're not moving forward so it's the writings job now they're not a fringe on game bar actually be gameplay it's one of the first things we prototyped check out that exclusive reveal of garbage but uh because the game what we're gonna is in first person and it's an ideal world we'd absolutely

never take control away from the player we need tools and again these principles do interrupts dialogue in a realistic way because you can walk around with this thing that makes people talk so for instance let's say like Henry Lee Henry and Delilah are out and you're talking about a windstorm that happened last night because dialogue about weather is like a surefire way to be super successful but like mid-conversation the player discovers like a necklace of human fingers like on the ground it's not in the game actually sorry I occurred to me that if I put that in like two seconds like I'm like oh people might think you could find that in this game I'm just suggesting that as a weird thing there's no necklace of human fingers in this game furthermore I googled that to put it into the talk and then I realized that I showed people what I found that somebody would be sick in barf so you get a look at a picture that I drew instead anyway you find this necklace of human fingers

Henri better say like whoa there's like some finger jewelry here holy and stop talking about the wind because this has now become much more important we have to interrupt the dialogue naturally and that's like really scary to do so it means that I actually means things for how I write the dialogue it means I can't write super lengthy lines if we need to finish in order to be interrupted in natural fashion I need to be able to tie in the new dialogue to interrupt in concordance with the thing that the player has just found or if it's even just like on screen whatever event has triggered that we need all the dialogue to exist in an organized and discrete chunk so it can be rearranged the speech manager needs to know exactly what's happening in the game at all times and sometimes the code of the conversation actually needs to change so like I have to write short snappy back-and-forth dialogue and I have to save longer monologues for times and I know stuff can't happen which is super

rare and then actually voice direct lines and tighten accordance with the possible like tones of the story sometimes duplicating conversations pretty impose a specific event but the thing that allows any of this to work is the system and methodologies used to organize the written content of the game so we actually took major cues from elan reskins 2012 GDC talk about Val's dialogue system there are aspects of a they don't quite work for our game because it's really dialogue choice heavy excuse me but the principles he outlines in the talk are like it which is available outside of the ball you can just go find it they've been in like an invaluable cornerstone to the thing that we're trying to build so yeah I'll share with you where we're at and the problems we're considering and some of the ways we're thinking about solving them to really over simplify the process she just really go watch the talk each of the line is triggered each of the lines is triggered by a singular event and is aware of the appropriate context at what point in the game is allowed to be triggered the dialogue system and the

event manager worked together to find the most pertinent line in the database and deliver it to the player at the most appropriate time so we want to get we want a very like user-friendly looks like I guess I'm the user and I'm an idiot so I need to like have a very user friendly way to generate and organize dialog contents and story moments because Firewatch is like is not procedural it's more systems driven than some of the things we've ever worked on but it's not procedural at all it's not a puzzle game like content is it being generated by a robot the game is gonna rely on lots of responsive content succeed and we're super small team so we have to have to be able to generate that and implement it really quickly and we've established goals for high right and now we're developing this tool to help me get there for now I work in a relatively freeform way and then chop up the individual lines so they can be put new database and right now I'm Patrick Ewing who's a programmer we're working with at Campo Santo and I are trying to flatten the process from written dialogue line

to implementation in the database into the game so to put it like as simply as possible we're trying to consider what are the fewest amounts of button presses it takes to like write a line store all of its logic and data record it implement it and tinker with the timing to ship in the final game we're trying to solve this really complex problem which is compounded by a realization that we both had which is how these tools work is going to dictate what comes out of the screen not just the pipeline it means fundamental things for the nature of the games cadence and tone the dialogue itself so if we're gonna choose to develop a system predicated on world state and the possibility of interruption and to respond to lots of different player inputs then the content we feed that system the actual writing has to be compatible every choice you make about how the tool works alongside the game systems impacts what we're able to put into the system and therefore we're able to push out of the screen because it's the earth like again we need to make a much as much of it as

possible in this game you're gonna be talking to someone over a radio as you explore in most games when you write dialogue you know a character is going to be in a certain place at a certain time you literally bring a character with you anywhere so the dialogue has to reflect that and to do that we need like just a shitload of it so we struggle to consider flexibility against agility how feature-rich do we make the front end of a tool versus how good is it at making the exact game we're trying to make we err towards the latter obviously like we need to get meaning to make this one game but it's an inescapable lure to think well what if the tool could it doesn't just take time to build but that moment of wondering well what if this will also let us is going to impact the game you make the actual content if you give a craftsperson a brush versus a pencil you're gonna get two very different images and at the most basic level all we really need to do is rip like right down and store a bunch of data in a

database it's like we can easily use Excel we just organized a bunch of lines as individual events so the player does X or world state is why we just make thing happen so what we've struggled what we've settled on this far is creating a very simple interface that has as much search ability as possible because we want to be able to look at a line at a tool and give it a bunch of new requirements on the fly requirements as the stuff that says oh I can happen when this is true furthermore we know we can take a big bite out of the polishing phase of the game so it's gonna be amazing if we got to be tuned the timing of the dialogue right in the tool we want I focused on a couple of core features like this it's gonna help us debug quickly and shift the dialogue faster but I want to point out that decisions to create and deliver content a certain way to build a certain tool actually help us combat the hurdles that stand in the way to creating suspension of disbelief that I mention before long cutscenes start to go away because of our methodology bad dialogue timing can

hopefully be saw over in the tool we can respond to player choice more deftly we're trying to flatten the content pipeline the same goes for interactions story and cutscene density as lots of people who have always said like game design like game writing is game design its level design its programming it's like the most disciplinary like multidisciplinary thing that you can do I think at times and in my day to day game writing is actually like systems design plus tools development plus creative direction so I need to be aware that things can't change without impacting one another and if I have a clearer picture in my mind for what the final product is supposed to feel like I can make informed decisions about all these things like not as like a writer or director or a project lead or whatever but simply is just someone who's like making things and to pursuit of a broader and creative goal so that's it I'll be finding some of this useful hopefully the game comes

out as good yeah I hope it results in a cool game for us to play I don't think we do QA of these is that right but um yeah you can find me there if you have questions maybe we do I don't know doing that okay so thank you that's it [Applause]