A Long Dark Road: Blending Player and Authored Story in a Sandbox Survival Game

Very last slide of my presentation I have uh some contact information on there for you. So, my email address and and my Twitter account. Um so, feel free to reach out to me after the talk if if you wish to follow up on any of the subjects that I covered today. Um how many of you are familiar with The Long Dark? Oh, gosh. Okay. I guess I better be careful what I say then. Um Okay, how many of you have actually played it though? Okay. All right, good. My favorite new thing is when you ask people like in job interviews if they've played your game and they all say no, but I've watched a lot of streams of it, which is not exactly the same thing. So, I'm just going to for the sake of kind of setting the stage for this presentation, I I spent a lot of time talking about the like making comparisons between sandbox mode in our game and story mode in the game. So, I thought a good way to kind of refresh that would be I'll play I'm going to play our our launch trailers

for both. Um just to kind of highlight maybe the differences in how we present the two modes. And then I'll segue into the presentation. The reason I'm doing that is because I made the critical mistake of a game developer, which is I'm using new software in a live operation. So, I'm not good enough with Keynote to do a proper swap between video and and uh and my Keynote. So, I'm going to start with playing by playing the sandbox trailer that we launched. It's almost 2 years old now, so you'll since you're all really familiar with the game, you'll probably see some things that you know are outdated by now, but this is I think a good um way to establish how we how we think about how we talk about and how we present the the sandbox mode. Excuse me. Oh, shoot. This is what I was talking about. Sorry. One second. one second. I'm very sorry.

Or maybe I won't do that. All right, let's try that again, shall we? Better? Thank you.

All All so that's that's sandbox. And uh you know sort of the emphasis on the atmosphere, the ambient experience.

That's really the focus of the sandbox experience for us. And then by comparison I'm going to play for you the the trailer that we launched on August 1st when we launched the first two episodes of our story mode last year. I didn't come here to fight about the past. I need a pilot to take me somewhere remote. Someone who won't ask too many questions. Someone I can trust. Wait, questions? Astrid, are you in trouble? Are you going to help me or not? The power went out. Who knows the reason? People started to get cold, hungry. They panicked, started to talk crazy. Others walked here from the highway.

Some belonged here. Others were outsiders. Like you. Yes, see I was part of that world. It's a world of closed doors and things being off limits. But that world's over. This has all happened to me before. Soon we'll all be tested. We'll have to choose how far we'll go to survive. You don't have to get it, Mackenzie. You just have to do it. But I did. Choice I made. I did it for us.

He'll be back soon enough. And when he is, I'll be ready. Holy Wait. Hold your fire. They left her, didn't they? She needed help. And they let her go. This is how we make it right. Astrid!

All right. So, thanks for coming. I hope you learned a lot today and uh buy The Long Dark. All right, let's see if we can do this correctly. Does it look right to you? Yes? Okay, great. So, hi. My name's Raphael van Lierop. I'm the founder and creative director at Hinterland and the creator of The Long Dark. I'm here to share with you my thoughts and experiences around the challenges of blending two types of narrative, the authored story and the played story, a problem that's fairly unique to the interactive space, and in a genre not well known for its narrative potential, the sandbox survival game.

But before I get into that, I'd like to thank the GDC selection committee and the advisory board, and in particular Clint Hocking for inviting me to speak here at the main conference. This is my third time presenting at GDC, and it's always an honor to stand amongst and in front of such distinguished creators and survivors of game development. I'd also like to thank my team at Hinterland, some of whom are sitting in the audience today, back there. So, if there's things you don't like about the game, and without whom I wouldn't have much to stand up here and talk about. I'm going to start with the standard GDC caveats. I'm not a lawyer, and forgive me if this sounds like disclaimers against litigation, but really this is about creating a safe space within which to share information and ideas. This caveat allows me to feel comfortable in sharing my experiences with you today without the pressure of feeling like I'm trying to teach you or influence you in some way.

I'm really here to share the benefit of some lessons hard won. It's up to you then to decide if this information is of any use to you. If you wish to embrace it, great. And I hope it helps guide you to success in your own projects. If you wish to run away from it, also great. Perhaps I helped you to avoid some very painful experiences, or inspired you to blaze a much better path forward for your own project. We can't really talk about player versus authored story without first agreeing on what we mean by story. For the purposes of this discussion, I mean story as an architectural construct. It's a set of consecutive moments that are imbued with meaning due to their context. While game mechanics are the building blocks of the gameplay experience, that is what the player does in the game. Story is what emerges when you give these moment-to-moment experiences

meaning by how you think about them, how you feel about them, and how you talk about them with other people. But how you feel about your experience can be heavily determined by your sense of authorship over it. If I present to you a well-crafted story that you happily submit to, you'll come along for the ride and feel a strong sense of connection to what you have experienced. We've all played linear games that were extremely well done to the extent that we didn't notice, or at least if we noticed, we didn't resent the hand of the author guiding us along. If on the other hand, you feel that the story is reducing your sense of authorship over the experience, the mechanics of the game are at odds with what the authored experience is trying to convey, well, you end up with a dissonance that can ruin any sense of immersion you might have otherwise experienced.

And even though you may not be able to articulate exactly why, the game just won't feel right. The challenge in providing authored story as a framework over a game that is heavily biased towards player experimentation of the gameplay space, which is the definition of a sandbox survival game, is in striking the balance between the different types of story the player will experience. In The Long Dark, players experience story across a variety of layers. Each of these layers is supported by a kind of narrative scaffolding that is comprised of all the things you see, hear, do, and feel as you go through the game. This is just an an opinion, but I feel that players I believe that players who feel good about their experiences in The Long Dark enjoy the interplay between these layers. I hear snow and

wind wherever I go. Always. Uh between these layers and find them additive to the foundational gameplay experience, which for our game is really 90% about walking around and strategizing around your rapidly depleting resources. I also believe that some of our approach to isolating these layers into discrete experiences, our sandbox survival mode on one side and our story mode on the other, resulted for some players in a narrative experience that felt at odds with their expectations. To really understand that, we have to talk about the various layers of content and systems that generate what we talk about as the story of the game. And we also have to talk about our specific approach in The Long Dark to presenting these layers to the player over time. I want to talk about the layers of story

as I see them in The Long Dark. Maybe you can see these layers in your games, too. There's the world story layer, which is comprised of what the player sees around them as they explore the playable space. This is supported by the game's art direction, environment design, props, and the remnants of the game world's prior life before the player arrived to witness it. So, details like signage, narrative collectibles left behind, or mise en scène, this is the story that emerges from interacting with the physicality of the world. World story is heavily implied. It's a mostly ambient experience that players take in and mostly without thinking about it. As implied story, it doesn't present itself as such. It's more about look and feel, and in some cases it will be described as the world's personality.

An example of the world story would be our portrayal of the town of Milton in episode 1 of our story mode. You can tell just by looking at it that not all is well. You can tell the world is desolate, run-down, isolated, and decidedly not modern. Although you can also tell that you aren't in the 18th or 19th century due to details like power lines, solar panels, and cars. For anyone paying attention, these details combine to create a sense of the game world's prior life. They serve to draw you into the world and its mysteries. There's also a system story layer, which is the story that emerges from the things the player is doing in the game. This is mostly about stimuli that force decisions, the player's manipulation of mechanics, user interface, or controls in trying to execute on decisions, and then the outcomes of those

decisions, always forced by the interminable march of the game's underlying systems and tuning, which conspire to create a kind of experiential heartbeat. By my definition, the systems layer of story is about memorable moments that exist at the intersection of the game's interactivity. Everyone who's played The Long Dark has a story about getting lost in a blizzard. I'm sure it's happened to most of you at some point. Nearly dying and then upon the brink of death, miraculously encountering a cabin that seemingly emerged from out of nowhere. Or a church. Contrary to popular belief, there is nothing scripted or intentional about these moments except that we want them to happen. These stories emerge from players recounting a moment in time that at its heart is a purely mechanical process. The game systems conspire to create this moment, much like a printing press

conspiring to produce a page of a novel. But it's the context the player brings to this mechanical process that turns it into something dramatic, giving it meaning. Oh, jeez. Sit back. There we are. So, here we come to a definition of player story. Player story is what you get at the intersection of world and system story layers. The player can use their intuition and senses to generate a mental model of what the game world is, what it might have been before they arrived in it, and then can use that information to orient themselves on a visceral level. Once they engage with systems through controls and user interface, they can start to establish their own presence in the world. They start to feel some influence over how the world works, how systems react

to their choices. It's a stimulus and response operation that increases the player's connection to the game world with each interaction. Through this process, the player begins to feel a strong sense of ownership over the game world and their interactions with it. They feel that they are shaping the game and creating their own story along the way. For some players, their player story is the most important expression of narrative. And it's a much more powerful story for them than anything we can create. Cute bunny, eh? Authored story is different from player story in that it is a narrative structure that is imposed upon the game systems and world by external forces, namely, the creator of the game. This tends to follow more traditional narrative lines such as those we're familiar with from movies or films. Sorry, or novels. Movies or novels. But its primary characteristic is that it is

something that is brought in from the outside, not something that emerges naturally through the confluence of world and systems. It's also something that can, for some players, interfere with the sense of the player story. Because in terms of hierarchy, the authored story rules over all. This can be mitigated in games that have extensive systems supporting player choice and then allowing these choices to modify authored story in a meaningful way. Some RPGs are able to do this, for example, or in games where players can more or less opt out of the authored story and continue to experience the game primarily along the system and world story layers. A lot of open world games like Skyrim or Red Dead Redemption, for example, allow you to do this. That said, most open world games still impose an authored narrative structure that even the most free roaming mechanics-focused players will

eventually bump up against. Minecraft is one of the few open world games that comes to mind that exists primarily on the systems and world layer and issues authored story entirely. But I think it's generally true that for any game that has an intentional authored narrative, this authored narrative supersedes the player's story. In the sense that even if you want to ignore it, you'll likely need to engage with it at some point if you want to progress. So, why bother with authored narrative at all? When done poorly, it can conflict with your system story. And frankly, it's a lot easier and more cost-effective to create experiences primarily through systems than through authored narrative content. Well, I think the truth is that traditional stories work for a reason. We're hardwired to understand them, and the context they provide is an important way of adding meaning to events which can otherwise feel kind of pointless.

And while there's certainly a type of player who prefers their own story to the one a development team might come up with, a constituency that's grown a lot over the last 10 years or so, this is essentially the Minecraft generation. A lot of players just can't be bothered. They feel lost without the benefit of some overarching storyline. And they take comfort in knowing that someone has put some thought into why they are there, what they are doing, and how it's all going to come together in the end. Any X-Files fans here, Lost fans here? Remember when X-Files first came out, not the garbage that they're putting out now, but kind of the first, let's say, five seasons that were really good, okay? Or Lost, anyone ever got into Lost big way? Yeah? Do you remember the moment you realized that the creators didn't really have a clue where they were going with the story? The moment you realized they were just kind of making up as they went

along, right? That was the moment you probably, like me, threw your hands up in the air and said, "Screw this." Why was that moment upsetting to us? Because an authored story has taken us along for a ride and it's a ride we're willingly submitting to because we think we're being taken somewhere interesting. So, you know, it better be going somewhere interesting. And if you stop believing in the tour conductors, well, it's done. The illusion is broken. The magic is lost. You see behind the curtain and there's no point going any further. I love that GDC is happening in the middle of all this awesome construction. To be clear, in this talk I'm talking about player story and authored story in a solo survival game, okay? Multiplayer is a completely different beast. Multiplayer games are almost entirely, in my opinion, about system story with the added context of playing against other humans rather than AI being the thing that provides meaning to the system story.

I think of multiplayer survival more like participating in a sport. Your story there is about your performance, close calls, near misses, dramatic goals or saves, and so on. I don't play a lot of multiplayer games, but I recall a lot of rounds of co-op Halo or Left 4 Dead that were tremendously memorable experiences because of the reaction from the other humans I was playing with. But that gets into a whole other realm of motivations and psychology around player story that we don't really have time to get into right now. Suffice it to say that for the purposes of this talk today, we're focusing on solo games. And speaking of solo games, it's time to talk more about the most solo game of all, The Long Dark. You're sitting in a lecture hall listening to a talk with the subject blending player and authored story in a sandbox survival game. So, you might start by asking yourself, what does this guy know about it? See all this gray hair? It's actually coming out.

That's mostly come in the last 5 years or so. Coincidentally, since I founded Hinterland and started on the game working on the game that would become The Long Dark. If you'll allow me to stroll down memory lane, and I promise you this is a journey with an intentional ending, I'd like to recount for you some of the history of The Long Dark's development and how certain choices along the way had a dramatic impact on our approach to player versus author narrative. And I might just spare you some gray hairs in the process. My first concept of The Long Dark was a tablet game called Survival Story. We built a prototype on iPad, which let you navigate a simple top-down map and select points of interest or locations to explore, at which point the game would transition to a small 3D scene the player could move around in. It didn't take long to realize that this form of presentation would be too limiting for an experience that was meant to be 90% about immersive exploration, so we quickly shifted our focus to creating something for PC and Steam.

This was back in 2013, and you have to remember that Steam, while impressive as a digital distribution platform, didn't feel quite as omnipresent as it does today. Everyone, it seemed at that time, however, was developing games for iOS. It was a new gold rush. So, to step away from the platform that seemed to be ascendant for the more pokey-feeling PC felt like a big risk. Don't forget, this was still the era when analysts brazenly predicted the death of the PC. Still hasn't happened, by the way. In any case, PC felt like the felt right for the kind of experience we wanted to deliver, and Steam meant we could find an audience without a publisher. And satisfying these two needs was enough to merit this way forward. Most of the team at Hinterland, back then as today, came to Hinterland after long stints making games in the AAA side of the the which meant we didn't really know anything about social media.

Streaming wasn't really a thing, and our primary experience of community was the occasional piece of fan mail, like literal mail, like physical stuff they send in the mail, that would get sent in and posted on a bulletin board for all of us to see. The point here is we were building A Long Dark in an environment where community was becoming a thing. Nobody really knew what it meant or what one was, but we knew that we needed one. And a lot of of studios started the creation of their communities using, well, Kickstarter. And we did, too. We launched a Kickstarter campaign, experienced 30 days of terrible anxiety, and in the end we raised $250,000 and gained a community of nearly 7,000 backers. And this was amazing for us in our first wonderful experience of being front and center with our community, as opposed to interacting from behind the shield of the marketing and PR departments that we were accustomed to from our AAA days. And I think we got a bit excited,

maybe even a bit addicted, to the idea of having fans and feedback for the things we were doing. Our Kickstarter was for a heavily story-driven survival experience about a bush pilot named Will Mackenzie who'd crash in the Canadian wilderness in the aftermath of a mysterious geomagnetic disaster. Sound familiar? And that disaster would render all of our technology inert, thrusting us into a new dark age. And in the throes of a harsh Canadian winter, Mackenzie would have to struggle to survive while managing relationships with all the survivors who were upset at starving and pissed off wolves and not being able to check Facebook and stuff. So, with our fresh community in tow, we embarked on work to take our Kickstarter prototype and flush it out more with flush it out with more mechanics and content, with the emphasis being on these intriguing post-apocalyptic story moments we'd be delivering in these amazing 2D animated scenes.

What we quickly realized however, was that the core of our game, the systems layer, really needed more feedback. We'd released an early version of our gameplay systems and world layer, which we referred to as our sandbox, hitting hinting at its non-narrative roots to some of our backers and received our first playtesting feedback. And much like the outcome of our successful Kickstarter campaign, we were heady with having direct access to our players. We wanted more feedback. We also noticed a trend with some of our Kickstarter backers. Despite giving them an early build to play, many of them came back to us saying they wanted to wait. They didn't want to experience the game before it was finished. Specifically, they didn't want to experience the game before the story was complete. So we asked ourselves, how can we get more players more feedback, but in such a way that we wouldn't spoil the their first time with The Long Dark by making them go through

the clumsy first draft of a story experience? I then landed on a solution that would forever sunder our world and systems layers from our authored layer. I said, "Let's separate the mechanics from the story, release only the mechanics in one region of the game world, leave just enough info to provide context for where the player is and why they're there, but keep all the rest of the real story until it's done." The intention was to generate more player feedback on systems while holding back the story for maximum future impact. We launched our non-narrative sandbox mode on Steam early access on September 22nd, 2014, and it it kind of took off for us. In our first 3 months, we had 250,000 players. In our first year, we had half a million. By the time we launched our first episodes of story mode last August, nearly 3 years after launching our

sandbox on early access, we had over 1 and 1/2 million players across Steam and Xbox One. Today, we have over 2 million players across all platforms. And we still have three remaining episodes of our story mode to go. I won't lie. We didn't anticipate how big our sandbox experience would become. We didn't understand how the growth of the sandbox player community would essentially dilute our focus from the game's story mode to that of trying to support the game's the needs and requests of our rapidly growing community. And while our original Kickstarter backers may have fallen in love with the game's premise and narrative aspirations, our early access backers were buying into a non-narrative survival sandbox. Sure, we promised all our backers they'd get all five episodes of our story mode for free with their purchase of the game on early access or for backing it on Kickstarter. And some percentage of our early access players were certainly anticipating the game's story content, but the truth is that the majority of players who fell in love with The Long Dark fell in love

with the game's sandbox mode. They fell in love with the game's world and systems layers. They fell in love with their own player stories. And thus began the struggle, the light side dark side conflict that continues to influence and govern the thoughts in many corners of our community today. What is The Long Dark really about? Is it about your stories? Or is it about my story? The author's story, the story I wanted to tell in the game. Well, the good news is it's about both. But as the sandbox experience continued to gain momentum, I became increasingly frustrated by how much narrative ownership we were needing to sacrifice to keep the sandbox engine fueled and functioning. While the early sandbox regions were intentionally generic, we started hitting against constraints. The more world and systems we built to sustain the sandbox, the harder it

became to hold on to content that would be fundamental to making the game's story mode stand apart from the sandbox. Why was this important? Because we wanted to be able to launch twice. Once with our sandbox and once with our story mode. We didn't want to lose out on the chance to do a big reveal of our story content. And we wanted to ensure that by the time we launched our first episodes, people would still give a In order to accomplish this, we needed our story launch to essentially be a whole different experience. And the more we built up the sandbox, the longer it took to build up the story mode to the corresponding degree. Because, well, story was our idea. It was the reason we'd made the game to begin with. It was I wanted to present the story of Will Mackenzie and his estranged wife, Dr. Astrid Greenwood, and share their complicated relationship with players against the backdrop of a harsh wilderness survival in this post-apocalyptic aurora-illuminated Canadian landscape. I wanted to create an epic survival

story with a decidedly Canadian context and a new kind of disaster, a quiet apocalypse, that would be thoughtful and beautiful. But many of our players were happy to simply avoid freezing to death while they tried not to get eaten by wolves. Many, many times we heard the refrain, and we still hear it today, "I don't care if they spend any more time on story. I just want more sandbox updates." Maybe some of you write that. It's okay. In order to maximize the impact of our story mode launch, we kept the game world intentionally generic. Almost as if we wanted to it to stand as a placeholder for what the game would become at full launch. Regions took on totemic names like Pleasant Valley and Mystery Lake. Items had simple names and iconic labels that left as much as possible to the player's imagination. There was no readable text in the world, including on signage. We kept our world story vague and almost impressionistic

to allow maximum mileage in player story. We'd do as much as possible to influence the player's experience of the world and its history as little as possible. But in this was my fatal flaw. Well, not fatal cuz clearly I'm still here, but if I was a character in a novel would be my fatal flaw. Despite this intentional genericism of world story designed to support as much player freedom and ownership of the game experience as possible, I didn't really revise my internal definition for what the authored story of The Long Dark Wood or should be. Well, that's not entirely true. I'd actually made it way harder on myself and my team. I decided that it wasn't enough to just tell Mackenzie's story. We should also tell Astrid's story. And we shouldn't tell the story in such a way that the plot remained the same whether you selected Mackenzie or Astrid, a gender-based selection that would have no real impact on the story the player would experience. The Mass Effect approach. We would go all the way and create two entirely different stories.

Yikes. Back when the game was called Survival Story, the narrative would be delivered primarily in these motion graphic style cinematic scenes. You can see remnants of the style in our first episode actually in those flashback sequences in the hangar and in plane crash. With basic dialogue trees allowing some fairly simplistic player choice logic. And while the world would feel relatively open, the narrative structure would be pretty linear. Much more linear than a truly open world game. Because back then, Survival Story wasn't really an open-world game like The Long Dark would become. The point here is, I was still thinking about the structure of the story being more linear as it was in its initial iterations of the project's early days. And that failure to update the narrative structure to suit an increasingly open and free-form player experience created a tension and conflict in our player experience that persists today. We were also faced with the fact that

our game world, in the pursuit of supporting the thriving sandbox mode, had become far too big for us to support with one holistic chunk of mission and narrative content. Although I like the episodic structure for how it mapped to television-style writing, that felt like a great fit for our game, we ultimately had to find a way to break up our story into smaller pieces because our game had just become too big. So, when we launched the first two episodes of Wintermute, our story mode, players discovered an experience that was well, much more linear than they had become accustomed to having in their sandbox mode play. And not at all following the same general main quest, side quest, directional-less wandering paradigm that games like Fallout and Skyrim had trained them to expect. Add to that, they'd had 3 years of player story knowing that the authored story would come along someday. 3 years to think up all the ways in which the story could go. 3 years in which to create their own ideas of what the story of The Long Dark

should be. And while that wasn't something we could really live up to in some cases, and even for the people who may not have had such stratospheric expectations about the game's narrative mode, imposing a strictly linear mission and narrative structure on what for many of them was a vastly open-world survival sandbox, well, this just didn't work. It wasn't what they were expecting. It wasn't how other open-world games blended their player story and authored story. Essentially, we had gone against the true nature of what the game had become and forced an adventure game narrative structure onto what had become a very open, very free-form sandbox experience. Now, not all was lost, of course. While some people were frustrated with the new constraints, others embraced the goals and context provided by the game's narrative mode. While some players found that the game could never live up to the stories they had made up in their own heads, others found it a pleasant diversion from the game's sandbox mode, something to jump into as a kind of palate cleanser before moving back into a challenging 100-day

interloper game. Still others found the story mode provided the game with a necessary narrative momentum that the sandbox mode was sorely lacking for them. The story gave them a reason to play. One thing we've realized since we launched our episodes last August is that this dissonance between expectations set up by the game's sandbox experience and what we delivered in our story mode were largely a function of our lengthy early access campaign. We found that players who've jumped into the game since we launched our 1.0 with Wintermute actually play the story episodes as a kind of tutorial in preparation for the game's more challenging sandbox survival mode. This works as intended since the story mode not only introduces mechanics in a safer context, thus softening the hard entry into our permadeath survival mode, but also allows players to take all the world layer context from story mode with them into survival mode. Issues with mission linearity linearity

aside, I suspect that if we'd launched our story and survival mode simultaneously rather than with a 3-year interval between them, this perceived disconnect wouldn't have really been an issue. The other benefit of launching the first two episodes of our story mode is that it allowed us the opportunity to reset expectations about the world layer of the game by adding more specific information about what happened before the player arrived, essentially the world's prior life. As if to reinforce this, the first region in episode 1 was the town of Milton. The first major region in The Long Dark to have a non-generic name. We poured a lot of energy into making Milton feel like a real place. And it set a new standard for us in the level of realization and verisimilitude we'd achieve with our locations. We also took the opportunity to add some life to the world layer by leaving narrative collectibles behind. Notes that would again fill in some of the blanks regarding what had happened to the world of Great Bear Island before

you crashed your plane in the aurora-filled night sky. One way in which we attempted to unify the system story and authored story layers in The Long Dark was through the inclusion of our trust system. The idea here was that various predetermined narrative and knowledge breadcrumbs would be unlocked based on your gaining trust with survivors in the world. These breadcrumbs could be background on world events, map locations to supply supply caches, crafting blueprints, or even skill knowledge that would improve your capabilities as a survivor. These gameplay elements were contextualized with story moments. Typically dressed up with lore or backstory to provide at least a thin veneer of meaning outside of their value to the systems layer. In order to gain trust, however, the player has to provide that NPC survivor with supplies, resources they can't go out on their own can't get on their own, I'm sorry. And gathering these resources requires the player engage with the game's core exploration survival systems

loops. It's similar to the relationship mission progression has to narrative content in most games with a story. In a mission, you are provided with narrative context, a reason for why an NPC needs something from you. You then engage with game systems in order to achieve the stated goals, and your reward is typically some gameplay benefit, loot, progression, etc. And then also the unlocking of the next beat of narrative context. That's not unique to The Long Dark, that's just a general structure. Trust plays with the same structure, but streamlines it to its very essence. The context is simple. The goal is often something that the player needs, and in the process of trying to satisfy the objectives outlined by the narrative con- context, the player engages in the core systems loop, which then produces new opportunities for player story via the systems interactions. Unfortunately, in our current implementation, many players found the trust content feel very fetch-questy. In the context of a survival experience, it makes sense that NPCs are often

weaker than you, which try to get you to take risks on their behalf. And in a world of resource scarcity, the challenging of challenge of obtaining food, fuel, and other supplies to help ensure someone else's survival, potentially at the cost of your own success, creates an interesting tension. And it's thematically very relevant to our world. It's relevant to our game world, but I guess it's relevant to the real world, too. It was also a way for us to differentiate between the survival mode, an ultra egocentric experience, where you only have to worry about your own needs, to the challenges of having to consider the balancing of your needs against the needs of other people you encounter in the world. Many who need your help. How far will you go to survive? It's probably a testament to how systems-focused trust felt, and the fact that we were that we treated it individually from the rest of our dialogue, narrative, and mission content, that it presented as a thinly veiled progression system. In future iterations, we're making trust more integrated into the core experience of interacting with NPCs, and I still believe that as an experiment in

blending player story, system story, and authored story into a single system, trust netted some pretty interesting results. While the decision to split the game into sandbox and story modes allowed us to confidently launch our early access campaign without fear of spoiling the story content for our players, and to be clear, I think that was the right call given that I don't think heavily narrative games generally do well in the early access model. I think we ended up breaking with open world conventions in doing so. We created a split between our more heavily narrative focused experience and our more heavily systems focused sandbox experience. Whereas most open world games allow the two approaches to coexist, leaving it up to the player to choose for themselves what type of content they prefer to engage with. There have been some sandbox survival experiences who have managed to introduce their narrative mode while still in early access. The Forest and Subnautica come to mind as examples where this has been done well, and both those games have been able to integrate their narrative modes directly into

their core open world sandbox. For our game, however, we found that by the time we had a strong handle on what an episode for The Long Dark would look and play like, the game had already essentially become well, two games. The strongest expression of this difference is in the way we handle player failure and saves. In story mode, you can save anywhere, and failure can more or less be remedied by reloading from an earlier save point. But survival mode is permadeath. If you die, we delete your save. And the risk of the game's permadeath nature is, frankly, one of its core strengths. The danger of loss is what lends meaning to the system story in survival mode. Combining story and survival modes into one open world experience would have unified the two approaches, but removing permadeath from survival mode or adding it to story mode would destroy the spirit of each experience and, frankly, result in a worse game for everyone. So, this reunification is just not possible for us given how our game has evolved over time. So, while I think that our approach has taken us outside the standard

conventions of open world games, we're committed to making it work for the remainder of our five-episode story mode. That said, this doesn't mean we're happily standing still. One of the things you quickly learn about making sandbox survival games in an open development and community-focused environment is you're only as good as your last update. And after over 4 years of highly iterative open development on The Long Dark, the team at Hinterland is now heavily optimized for this particular development methodology. In other words, we can't leave well enough alone. So, we're working on several things to strengthen the relationship between player story and authored story in The Long Dark. Some of these are a bit spoilery, but hey, it's GDC. You have to get something for your money, right? We have an episodic story structure which does introduce some useful constraints for us. We can present a series of narrative nodes within a specific physical area. Keeping in mind that the survival mode version of Great Bear has eight major regions, four transition regions, adding up to over 50

square kilometers of terrain to explore, which makes it bigger than Skyrim. Episodes 1 and 2 combined only cover about 20% of this world, and the world is still growing. A lot of our narrative and mission structure for the episodes was actually designed by necessity to work around the design goal of constraining the player to specific specific areas before letting them move on. This is a common technique used in open world games, but in general, the constraints are softer and the boundaries are a bit less binary than in our episodes. Our goal is also to make our story delivery feel more like a television series than like a film. So, we have hard openings and cliffhanger endings in the episodes to hopefully draw people back in after they complete each self-contained episode, which to be clear are not Telltale-sized. Our episodes are between 5 and 10 hours long. In any case, one of the things we're doing to unify player and authored story a bit more moving forward is making our mission structure less linear. We've actually gone back into our episode 1 and 2 mission scripting and opened things up significantly, so that you can now complete most of the major objectives in whatever order you

like. And in some cases, you can completely opt out of mission threads that were previously mandatory. We're also redoing the bear hunt for anyone who might want to know that. In order to support that this increased openness, we also had to move away from our previous cinematic presentation style where we book ended dialogue sequences with non-interactive cinematics towards something that can support enhanced player choice. We've been working on a new kind of dialogue mode for The Long Dark. Something that does a better job of integrating the system layer of the game. This thing. And it's contextualization within story moments into interactions with NPCs, and also an interface style that itself represents the nuance around what it's like to interact with a character who might be scared or mistrustful of you. As you can see from this example, topics sort of float around NPCs like dust motes, and their position relative to you, their size, color, and behavior all indicate something about their context. Larger fonts mean these are subjects the NPC really wants to talk to you about. Locked items suggest topics that are

available if only you first satisfy a gameplay requirement. We use question marks to indicate that this NPC has more to say to you in the future. And the speed at which some topics move, or the fact that they may fade in and out, making them hard to select or in a conversation, reflects on an NPC who might be trying to be evasive about something. This approach is not This approach not only blends the systems layer and authored layer much more organically, but it's also scalable and dynamic enough to be reactive to player choices, which supports our goal of opening up the mission and therefore the narrative structure. Well, I certainly don't recommend that you go back and spend a lot of time and money redoing content that you already spent a lot of time and money on. Content that a lot of your players have already experienced and may not want to experience again. But we see these improvements as important work for strengthening the foundation of our game so that we can build the story future story episode content such that the player story and authored story coexist more cohesively.

So, after all this, what comes next? Well, as far as this presentation goes, I'd like to leave you with a few high-level takeaways, things for you to consider in your own projects, and then with whatever time we have left, open the floor to your questions. For a lot of players, and this is particularly true if you have an open world-ish or sandbox-ish game, their story is a lot more important to them than anything you might come up with. Make sure that you create tools that allow them to influence the game world in a way that gives them a sense of ownership. In The Long Dark, we often have players post lengthy survival diaries that narrate the moment-to-moment mechanics of their sandbox experiences. To a lesser degree, players share their experiences of the story mode, but there's just less for them to own there that other players aren't also experiencing. Our survival sandbox has become an engine for memorable moments, which makes it appealing for streaming, itself a kind of player storytelling. And we have an in-game diary that encourages players to keep notes of what they've done each day, or remark on where they may have stashed supplies or encountered

obstacles they'd like to come back to when they're in better shape. We also think about how to create interesting choices within our mechanics, things like our rabbit harvesting interaction, where you have to make a choice about snapping a cute bunny's neck or releasing it and going hungry yourself. These are relatively small investments from an implementation standpoint, but they can create tons of animated discussion amongst your community. A lot of it in public places where other people's curiosity might drive them to check out your game. In this case, the rabbit harvesting also becomes core to the player's story, and the choice to harvest or release, as in BioShock, is something players will reflect deeply on, strengthening their bond with the game. The point here is, sometimes small tools dedicated to player sharing can go a long way when they are in support of a game with a high index of player-generated story moments. A lot of your players won't be satisfied with only having player story. Authored story still has a very important role to play in setting up overall context and providing import necessary momentum and motivation for player activities.

If you give players the opportunity to opt out of the authored story from time to time in ways that feel meaningful, then the two approaches to story can complement one another nicely. This is in fact the true strength of the open world genre. And if you don't have the budget for cinematics, VO, facial animation, etc., find a more cost-effective way to present your narrative moments. There are many great examples of how to do this across the spectrum of independent games. So, don't abandon authored narrative because you're afraid it will be too expensive. Consistency is more important than hitting a particular presentation style. This is less of a note on narrative approach and more about general community engagement or setting expectations. If you're making a sandbox survival game in an open development environment like early access or game preview, be very clear about the kind of game you're making when you talk about it with your community. We spent nearly 4 years publicly making a super open free-form survival sandbox game, and at the same time made a narrative mode for it that we kept very much in secret. Even though we tried to explain that our story mode would be different from sandbox, for fear of spoiling stuff, we never really articulated exactly how our

story mode would differ from sandbox. It might be better to give up some of the surprise factor and just be more forthcoming about the exact nature of your game structure, assuming you were holding major stuff back from your community. I think in general, you can get a community on board with pretty major changes to a game, provided you ease them into it, which in the case of our story mode we did not do. I would never suggest that you anyone design a game specifically to make it more streamable, but if your game naturally lends itself to this type of player storytelling, then do whatever you can to make the most of it. That said, for offline games like The Long Dark, the most important storytelling tends to happen outside the game, in the community itself. So, try to provide some in-game tools to facilitate player players recording their own experiences. A journal-type system is a good place to start. And outside the game, make sure you dedicate some safe community space for the sharing of fan fiction and player diaries. Not only will this serve as a stage for your players' creativity, but their excitement about their own stories will be infectious and encourage others to explore your game.

Well, I don't think we've necessarily solved the problem of better blending of player and authored story with The Long Dark, at least not yet, and I don't pretend that I've provided the solution in this talk. The balancing act between the imposition of authored story on player story is one that many open world games are still struggling to solve. My theory is that it's not really possible to find a balance between these elements that satisfies all player types, so that providing as much freedom as possible is probably the safest path forward. In this way, our solution of creating two different modes, one specifically targeted at systemic anecdote creation and one specifically targeted at a more traditional narrative, might be a viable alternative to the vast blended open world experiences that are the purview of the Skyrims and Grand Theft Autos of the world. Because ultimately, whether your game has discrete systems designed to support player or authored story, or it doesn't, the player's experience of your game and how they talk about it with other people, whether it's in streams, in forums, blogs, or in social media, will

be the most powerful story they experience. Thank you. So, we have 7 minutes for questions. Anyone who knows me knows that I'm a terrible rambler, which is why I had to read off my slides, or else I would have run out of time a long time ago. Go ahead. Hi. Hi. This was a really insightful talk. Thank you very much. Thank you very much, too. If the if it would have gone the other way around and you'd have had um narrative authored story first, Thanks to everyone who came. Sorry. for however long, a of years or something, How do you imagine the challenges would have been different? So, the question is if we had done it the other way around, so started with authored narrative and then done sandbox, how would the challenges have been different? I mean, I don't think I think that's sort of what I was trying to get out with the the layers idea is that you I don't know that in a open world or

sandbox game, you can't have a like a an authored narrative layer without a systems layer, a world layer, etc. etc. So, I can't off the top of my head think of of an example of a game that does that. Maybe you can. Um Yeah, if you if say if you'd gone like um you can only allowed the player to experience it through the authored story Mhm. mode. be more like an adventure game? Yeah, and then and then like it would feel like a really linear story. Oh, okay, I've finished the story. And now you start to enter players, oh, we've got this other sandbox mode Right. Yeah, I think what would have happened in that case is then sort of the dis- the differences would be even more stark like because you would literally would have a very linear game and then this other open world game kind of bolted onto the side. So, even though we did we do sort of differentiate between the two things, even our story mode for a lot of players, it's not actually as linear as I portrayed it here, I don't think anyways. I think there's still a lot of freedom within the constraints of each episode. And a good example of that is I remember when we launched we had a a a a

piece of feedback in our community from a player who said uh you know, I find episode one so boring. Um I spent 90 hours wandering around Milton and there's just like nothing to do. Right? And and we're like, yeah, but it was only designed for you to be there for like five or six hours and then it just highlighted that I assumed that this player was someone who had come to us from the sandbox mode where they would maximize their time in a space and optimize their resource use as much as possible ignoring the fact that there was a whole story going on in episode one that was trying to kind of give them a momentum forward. So, to answer your question directly, I I'm not sure we would have ever done that, but I think if we had, we would have ended up with a two very different games, even more different than they are currently. Great. Thank you very much. Thank you very much. Hi. Hi. So, thank you so much for the talk. That was That was super interesting. Thank you. I have a a question about it. So, it seemed to me one of the things you were pointing towards in player story is the importance of novelty. Is the idea that that you as a player can always generate new content for

yourself and new story content for yourself. And I was wondering if you think that may be the like the key hook in making players so enamored of player's story and how you think it's what ways you think you can best push that into the authored story and get the same sort of novelty hook, even if it's a story with which the player is somewhat more familiar. It's a great question. I'm going to try to rephrase it, but you might have to help me out. So, work with me here. So, the novelty of the player story is what makes that experience interesting. What's the your asking how we can Yeah, so that's my question. It seemed like your your part of your argument for why it makes it interesting is the the novelty, right? If that's the case, how can you best push that into the authored story? Yeah, I'm not sure if I would have used the word novelty myself. I think of it more from the perspective of ownership. So, I think there's just a natural feeling and that's kind of why I tried to make the note about, you know, if a if a story is done really well, you kind of go along for the ride and you

don't mind and you sort of that's why I use the word like words like submit. You submit to it because you're giving up some freedom to do that. In the sandbox mode, where it's more about player story, there is there are no constraints other than the mechanical constraints of freezing to death, etc. etc. So, I think in those cases the player players that really align themselves with that style of game derive much more enjoyment out of those little moments that happen kind of along the way as they're experiencing the game systems. So, I think what we what we tried to do with our story mode was exactly what you're suggesting, which was we tried to use the mechanics layer of the game, the system story, you know, as the uh I won't say filler, that's the wrong word, but basically the narrative or the story layer, if I put it the other way, the story layer was really providing the context and kind of that momentum and the heartbeat of what was going on in the world, and the systems layer was what was providing the moments for those, you know, moment-to-moment experiences that would hopefully be memorable for players. And so, that was really our intention with that with that survival mode. Sort of echoing what I said over here,

I think where um where we ran into problems was just that, you know, when you have a 50 square kilometer world and you can wander aimlessly forever, I mean, we have players that play for, you know, 1,500 2,000 hours kind of in the sandbox world, even putting them in an area that's 10 square kilometers suddenly feels very restrictive to them. So, but it but anyways, to your question, I think we had tried to do that with with our story mode. Um and I actually think that that's, if you think about any open world game that's successful, I think they all are doing that. They're providing that over our overarching narrative to give the player reason to continue, but then really it's the moment-to-moment moment experience of the game, which is what kind of brings them back. And those are the things that people tend to talk about. They don't tend to talk about like, "Oh my god, remember that awesome cinematic that happened where that character said that thing to that other character?" It's really more about like, "Holy cow, do you remember that moment that that thing happened in the game that I did in the game, right?" Does that answer your question? Yeah. Not not really? Feel free to get in touch with me afterwards. It's pretty nuanced stuff, so. Yeah. Okay. But thank

you. Yeah, thank you. Okay, thank you. Anyone else? All right, you guys let me off easy. All right, thank you very much for coming. Oh, we've got one more here, I'm sorry. I shouldn't have spoke too soon. Hi. How similar is the story mode as it launched to how you envisioned it at the time of the Kickstarter pitch? I mean, in terms of realization, we were able to achieve much more than what we'd expected to achieve um in our Kickstarter pitch. We had a fairly small budget when we did that campaign. Um fairly modest uh ambitions for what we would do with the game. And um so what we ended up delivering is I feel um at a much higher level of quality, a much more impressive level of realization in terms of facial animation, character realization, and the music, and the VO, and everything that's going on there. So certainly in our original intention, things were going to be much more modest in terms of

presentation. So we were able to not only provide a much bigger world that had a much richer set of systems to play with, but also the actual narrative content itself was at a higher level of realization. So but but the actual underlying story or the intention of the game pretty much didn't change at all from the Kickstarter to launch um with the exception of adding the Astrid story. So in the original sort of idea for The Long Dark, you would just play Mackenzie's story kind of all the way through, and Astrid was there as a as a character to kind of help provide motivation for doing certain things. But it but when we launch our third episode, you're actually going to be playing as Astrid. So you're going to see the sort of the experience of The Long Dark from a completely different perspective. She's a doctor, she's in a different part of the world than you are. So sort of we'll show her part of the story as well and kind of go back and forth between the two characters. So that was something that we would never have been able to do um in our Kickstarter or like in the pitch that we had for the Kickstarter campaign, and would have been far outside of our reach. Uh I think it's not outside of our reach right now. Does that answer your question? Okay, great. Anyways, it seems like we were out of time. So thank you so much

for coming, and feel free to get in touch with me afterwards if you like. Thank you.