Designing for Non-Linear Story Discovery in Tacoma

[Music] uh first talk of the day as you can see is steve and nina here from fulbright talking about level design in tacoma hello i'm nina freeman um and i'm here with steve gainer to talk about designing for non-linear story discovery in tacoma which is a game that we worked on and released this past year um so i i am a level designer at fulbright and obviously i worked on tacoma tacoma for those of you who haven't played it here's a screenshot of it it's a first person 3d exploration game set on a space station in the near future it's about a woman named amy who's been sent to retrieve an ai from a space station called tacoma that has been recently evacuated for a reason that you as the player are not fully aware of in the beginning of the game

um and you play as amy and sort of discover what happened with this crew that left and why the ai is still there and all this stuff so the game is really about exploring the station to learn about what happened there recently and the story is communicated on multiple levels throughout the game both mechanically and through the environment which is what sort of this whole topic is because it's all sort of relating back to level design for us so we'll explain what some of our design goals were how they relate to level design specifically and what it was like actually implementing them throughout the development of tacoma from the beginning to the end basically so i'll pass the torch along to steve who will kick off our little discussion peace thanks nina um let me see can i angle that a little bit no i can't um so uh yeah i i was the um co-story lead and kind of the you know creative direction uh

force of the of the game and uh i come from a level design background so you know like nina was saying fulbright fulbright in the level design summit is a piece of level design jargon so like the name of our studio is a level design term so we have a very like level design centric kind of founders syndrome about how we tell stories and and what the focus of our design processes and how we relate to the player so on my side of the talk i'm going to talk a little bit about the kind of top-down direction and player experiential intent behind the game how we got to that from gone home and what we were trying to do with tacoma and how we worked together um to put that on screen um so yeah the our first game um at fulbright was gone home which there's a screenshot of up here um which was a game about exploring an empty house uh well empty of people uh house to find out about what happened um

to the people who lived there um it was a you know story exploration game a uh a walking simulator uh of of early vintage um and and it really uh focused on just the environment the objects in it and what you could learn from that and finding you know audio diaries that explain the story within that but going from gone home to tacoma we wanted to expand on what we do with these kinds of games mechanically and work more on how the player is involved with the moments in the story and how they engage with finding the story actively in a way that we didn't in gone home we wanted to bring the player closer to moments in the story kind of as they happened while also allowing them to have the full freedom to find every detail they're interested in like they had in gonhome so as part of that we were really

heavily inspired by um immersive theater productions like sleep no more uh where the audience shares the space with the performers and follows them through the the performance space and kind of chooses their own path through these scenes that are that are happening um something that is uh i'll talk a little bit more about sleep no more it's a it's a performance that's been in production for like i don't know like six seven years now in um in new york and it takes place across three floors of a converted hotel i think it's actually a converted warehouse they made look like a hotel but um basically you as an audience member in this space and there's a lot of performers that are moving through it along their own paths and you as an individual can only follow one you know you can only be in one place at one time so the the performance is an hour long but it loops three times um over the the course of the night and so you as an individual can say like oh i followed these performers into this scene and then followed that guy last time i'm going to follow the lady

that was the other side of it this time and see more of the total but you're kind of building up partial information about what the entire performance is through what you choose to observe we wanted to take that core experience which we felt is really unique and valuable and convert it into a form that felt native to a digital experience and so for for us what that meant was that since this is a digital uh game instead of it just looping and you choosing what you're going to do on the next loop we could give you the ability to fast forward and rewind and pause and be in control of both the performance and your experience of it instead of just your experience of the performance so we wanted to give players the tools to understand the story as it happens by actually being able to have hooks into it um and

and what they are seeing of it and what it's doing along with being observant um basically to be a more active observer than we had um in in our prior games um but you know we didn't start from this as our set of core mechanics when we started working on tacoma when we started we we had um a much more traditional uh progression style um you know we wanted we started with a story and progression structure that was very close to to gone home you know here we have you know the first thing you do in gone home is move an object and find a key and unlock a door with it which is like extremely traditional we do have some lock doors and stuff in tacoma but it was more at the center of what we were doing in our first rev of the game um you know the this this progression style is derived from classic audio diary based level design um you know which so i worked on the um multiple people at the studio worked

on the bioshock series before we worked at fulbright and i kind of learned how to tell a story in a level from those games and you know i was very familiar with how you stage out spaces to place story breadcrumbs to have the player kind of move through a story as they move through a space and you know in a lot of ways i was probably too familiar with it because we said we should do that again and you know like that's fair right like i don't i don't know like it is a it's a it's a legitimate way to to make a thing but as we started to to do it um you know we we kind of realized that it wasn't speaking fully to what we actually wanted to do with the game you know in a in a mostly linear story game you use audio diaries um you know to match the player's flow through the space in gone home you're discovering this story of a young woman named sam who's a high schooler and kind of what's happened to

her in the year that you've been away she um moves to a new house which is the house you're exploring goes to a new school meets a another uh young lady at the school and they fall in love and so you know in the game this is the first part of um gone home this is the entry and the west hall of the house and so you know we arrange the plot points in the order that you'll most likely encounter them and it you know it's it it it can feel non-linear it at least is player driven and self-guided but it still is saying we're going to send you on a reliable path through these spaces that tell you a linear story and you know in its original state tacoma followed the same design approach the spaces were laid out kind of as a string of pearls and each point in it contained an isolated section of the story or a plot point or a moment and you moved through them and they were basically spent and then you continued

on your way and our first pass included these ar projections these 3d recordings that ended up being the core of the the finished game but they were much more isolated they're essentially you know visual audio diaries i think that we've seen these a lot in games um you know where you kind of are playing a sci-fi game and you walk into a room and a you know hologram shows up and you don't really need to look at it usually it's it's just you know it's it's visual flavor for an audio diary um but as we were doing that we realized that yeah it wasn't really different it wasn't really you know something that we hadn't done before it was just a sci-fi version of here's your audio diary and so you know with that version of the game we basically it showed us what we wanted to do with this you know element of what we were doing because it showed us what it wasn't doing um so you know the fact that these ar characters actually fit in

our kind of staid level design process showed that they kind of weren't having enough impact on the game because they they didn't force us to design it differently you know there were one or two places in the original version of the game where these ar scenes kind of spread out over more space and showed us the potential of like okay this person leaves the room and you can either stay in the room or follow them and you're kind of doing that that piecing together but it still was localized it was still in on the scale of just like you know kind of a 20-foot radius or you know 40-foot radius or something um but it showed us that we could do more with it if we thought about level design differently you know our levels needed to be reshaped to embrace non-linear interconnected ar scenes which means they needed to be designed for those scenes not be designed and then we put the scenes into them um you know

they needed to be just larger and more complex have more going on in them the scenes needed to be longer they needed to cover more area and the kind of area they covered needed to be different um it basically was the thing that forced us out of our traditional you know shock game audio diary um level design uh uh territory uh which was good but it also meant that we had a lot of new problems to solve which is also good except when you're trying to ship a game because it's hard uh it takes a long time um so you know our our ar scenes turned from these points where you would kind of encounter one character a couple of characters and and either two webs they had multiple characters that all had individual threads that were running parallel to each other and that like sleep no more you couldn't see all of at once in fact you can't see all of any of our central ar scenes unless you pause and rewind and move through them a number of times um

behind me and we leave that to the to the player to determine how deep they want to go into seeing the the the content of these scenes um but that was also good because that is in line with what we wanted the player's relationship to gone home to be which is to say there's a house here there's a bunch of stuff in it if you find all the stuff because you wanted to find all the stuff not because you know you felt like you had to or you're checking off a list or whatever and so similarly you know at the point that we expanded on what our ar scenes were it reinforced that central player experience in a totally different form so that's good but it took us a while to get there it took us a first version of the game that we were like oh this isn't what we're doing um to realize you know what that was so you know now when you play tacoma you see our version of what it means for these ar scenes not to just be visual audio diaries but to be the central

mechanic of the game the way that you discover the story and what the rest of the game is supporting as opposed to being a supplemental part of the overall experience um so that's our that was that was our um our direction right we worked together to discover the mechanics and what the requirements were for that central player experience but then the question became you know how do we actually build those spaces and build the scenes and do the work to support the player through that entire experience in the details not just at the high level and so nina's going to talk a little bit about that process of putting it on screen okay one thing i want to add to what steve was just saying which wasn't in my

slides but now i want to say it is that people other developers might find it interesting that in that process of going from more of a string of pearls approach to this like interconnected ar scene approach with the rewinding and fast forwarding that we ended up with we had a whole full version like of the game that we had built that we just like basically completely threw in the trash and that was like over like a year of work maybe it was like that we uh sometimes you you do pre-production on purpose yeah sometimes you realize you were doing it afterwards yeah exactly i spent a year on pre-pro without realizing so like this is all a big learning experience for us i feel like and yeah we really did like kind of end up starting largely from scratch when we found the actual mechanics that we felt like inspired by so maybe just a little nugget um that people might find interesting okay so i'm going to talk about like paper mapping gray blocking and all that sort of like implementation stuff so

what i knew when i started paper mapping um which i think it was like me steve and tynan i did a lot of it though um what i knew was that from this earlier version of tacoma the thing that was inspiring us that made us want to change the game so much were the scenes that were taking place in these more open interconnected spaces where characters in our scenes could walk around from room to room and sort of like pass each other in a natural way and that that felt like a pretty good structure for our ar scenes and i'll get into a little more detail about what that means um but i came into the paper maps knowing that and i knew a little bit about what the air scenes that stephen carla were developing and i knew that each part of the station would serve to house a specific character with a specific job um and thus would need the realistic facilities to support these functions so with that in mind i focused on creating non-linear spaces that would support

both the future ar scenes and sort of the more utilitarian aspect of like who are these characters and what are they actually doing there what do they have to do to realistically fulfill their job in their section of the station so it's like we have like our botanist has like his botany wing subwing and then we have like our engineer has like her area so sort of like job-based sections of the game that i was paper mapping early on uh so here's some text and not the paper maps because that's the next slide [Laughter] let's see oh it's because i didn't hit the button correctly okay so to get more into paper mapping here are some actual paper maps um i want to talk about why i think sort of our initial approach works so well with ar scenes that were eventually built with these and these are two of the early admin paper maps so the administration wing is like sort of the first big wing that you encounter in the game and it contains the operations

facilities and the um personnel the admin and operations facilities and it's called the personnel wing sorry so the earlier paper map is the one on the left and then the one on the right is sort of closer to what our final like grey block ended up being and what the final level is now in the game um so while i was paper mapping and working on that one on the left which is the early one i'd asked myself how the rooms would be laid out and what order they would be laid out in sort of in the space such that the people who live there could actually use them realistically so i think this is especially important because tacoma is about a bunch of characters that are living and working on a space station like in isolation with a small group so they're like literally living in their workspace and so how do i design a space that seems like it would actually be tolerable to someone who is in this sort of extreme situation so one of the obvious questions for this is like where does the bedroom

belong in this kind of space where do the personal spaces belong in this like primarily workspace oriented station indiana yeah can i interject um so something that so as far as the paper map goes something i think is a couple things i think are probably interesting are a so robert has done uh a blog post at some point about sort of like the the the uh the stages of of being a good level designer basically and how there's sort of like complexity and interconnectivity of space that you learn over time um and i think that you'll see that the one on the right is a much better version of that than the one on the left and it it both supports our scenes better because the spaces kind of have overlap the way the scenes do but it also just naturally makes it feel like a more believable place to be especially on a space station it's like they want to have a conservation of space so

as far as the fiction driving what we do we're like also we should bring it in so there's not wasted space because it's it's up um the other thing that i think is hopefully i don't know it's interesting to me um is that nina um before she worked at fulbright had only made like smaller 2d games and so this progression is also like nina learning to be a 3d level designer as this process was going on and i think kind of like it was very interesting to me to see your paper maps evolve as you kind of did a lot of revs of like i'll paper map it then i'll gray block it oh i can see why that isn't really what i was going for i'll paper map it again based on on that so some of the progression i think of the spaces was also like your like learning process as kind of becoming a level designer which was cool yeah it's hilarious i was looking through like the whole all my old like notebook of paper maps and i looked at the first one and it was just like a box and like i divided the box a couple times

it started somewhere it was so bad i didn't put that in any of the slides too embarrassing but yeah i did a lot of paper maps and one of the things we did focus on was yeah like cutting out stuff like cutting out as many hallways as possible and just like making our spaces as compact as possible to also support the realism and the utilitarianism of these spaces because like you wouldn't it doesn't make sense to have like such a sprawling space station like that wouldn't make sense for the company in the game to build such a huge space for the small amount of people who have kind of as we explore in the game like crappy jobs so it's like sort of a small very compact space and like the paper map process was super iterative um so like the time difference between these two paper maps are pretty wide like we did a lot of iterative paper mapping especially in the beginning um so like i was talking about the bedrooms so when we're thinking about like where a bedroom would go in like a workspace on a space station like this

you probably when you think about it you probably don't want it near the entrance of a wing because that doesn't feel very private right you also don't want to put someone's bunk right next to their office because like who wants to sleep next to their office like you probably don't want to do that and even though this is like a corporation building it i still don't think they would even do that that would be really evil we could have gone that direction um so we wanted to make sure that there was a clear separation between public and private spaces on the station given the fact that the station was built for long-term postings on it for these people so i iterated on these paper maps with this concept in mind and came to a decision that our levels should have the sort of hub and spoke structure so for example admins hub is the dining room so you can see it's sort of like that central space where all the other spaces branch off of it um and each space that branches off the central hub is separated to a different degree based on whether it's a workspace or a private

space so most workspaces branch directly from the hub without much space in between since they're meant to feel more easily accessible by multiple members of the crew who might need to use them for work so the conference room is right next to the dining room i have it circled right there and that conference room even has a window looking into the hub because a conference room is like an extra public space where like they're having meetings with many different people on the station um and the conference room is also like it's also sort of like associated with the dining room because it's like these sort of like public gathering spaces and this wing is really like dedicated to that because it's like administration this is where people come to meet with their supervisor evie to have meetings with her to go to the conference room to eat and so it kind of makes sense to have these places closer together for us and then sort of the opposite of that is the private space and in this wing i have circled here where the bunk is for evie evie st james she's the station

administrator so she lives in the administration wing and her bunk has this sort of nice hallway that's connected to that hub space but it has the separate hallway unlike the conference room which literally just has a door and a window separating it from the hub space so this is like an example of how we're like separating private space more deliberately and this kind of approach with the hub and spoke structure and separating private spaces really deliberately works super well for ar scenes because you can have conversations that intersect and break off in a natural way so for example you see a lot of conversations in tacoma intersect in the hub space of a wing where the characters would naturally be passing by each other so like say there's one part where it's like evie like comes out from her office and she's walking through and then she ends up in a conversation at the dinner table because she's like naturally moving from a private space to a public space and encounters a group there and stops to talk to them um

and then like with these more public spaces it like makes sense that they would be gathering there to hang out with each other right um so the open interconnectedness of these spaces worked super well because of that natural flow of people through them based on how they would be like living their lives naturally um so i have some more examples of this so we have sarah right here who's at the pool table in the administration wing talking to odin who's the station ai and she talks to odin a lot and it's great uh and so she's in this open space that the dining room overlooks so she you can see sort of like there's an upwards ramp going up there's the dining room above and then she's below with odin soon enough she's like she's sitting there playing pool and then nat actually comes over to the railing which i have her little yellow arrow pointing to nat and she like calls sarah over for cake which is happening upstairs this is kind of a neat moment because it's like you're maybe as a player

you're focused on what sarah is talking about to odin and that's your full attention is focused on that and then suddenly you hear this voice you know nat calling out like hey like look over here and you as the player will like shift usually and like look up see nat calling to sarah and then see another group of characters gathered up around the dining table and then that gives you a lot of information about oh okay so there's another part of the scene happening while i've been paying attention to sarah maybe i should rewind and go see what nat was talking to those people about so that's sort of a nice moment where the hub is really a space where you can kind of like see multiple nodes of conversation happening and be made aware to go check out another part of the scene later it's kind of nice you guys want to hear a dorky level design thing about that it's a scripting thing actually yeah so like you don't want to you know you don't want to be able to hear everybody from everywhere like a hard something that i think most developers don't really well anyway in this kind of game it's not very common for you to have to really worry about like a lot of overlapping conversation but we

did so it was like how do we make it so you can hear people a little bit from far away but it's not just stepping on each other some of that is our programmer hannah put in a system where if you're looking at a character they're louder which is nice if you look away from somebody they get quieter but that doesn't really speak to something like someone literally calling all the way across the space so you could do something smart and programmery to do that or you do like we did which is when she yells we script her voice range to be twice as far and then we put it back to its normal range when she's done yeah we do that in a bunch of spots because they're like yeah the far away conversations where people are like hey yeah so don't overthink it just turn it up uh yeah so this this worked out really well um and also if you think about it like it would have felt sort of i think i think it would have been less rewarding giving given our mechanics if we had for example had this playroom where sarah is

separated by a hall and a door from the dining room because then nat would have had to walk down the hall to get sarah and the player wouldn't be able to see the crowd at the dinner table right so this sort of hub structure where the spaces are interconnected and looking into each other gives us the opportunity to make these moments where you can see things in the distance like that which wouldn't work as well with too many like very separated rooms that don't look into each other so obviously not every ar scene is going to have like conversations that are always sort of interconnected like this and flowing in and out of each other sometimes there's going to be very private moments that we want you know to sort of isolate or that the characters would want to isolate themselves like the screenshot i have up there is when evie goes into her office which is a private space and she sort of like takes something out of a drawer and you can look and it's a photo of her sister and she has this sort of private moment um so it's for

these private moments where characters are being more introspective or having like a one-on-one conversation that it's useful to have the private spaces be separated like we did on purpose sort of for the utilitarian aspect it also helps us sort of design scenes that can have private moments because we have specific parts of the level that are designed to sort of make that possible um we dubbed in some of the sound of constructing these levels into the background of the talk for your pleasure yeah the dub the construction dub not subbed so we found a lot of useful ways to design ar scenes that made good use of the hub and spoke structure and there were there were still wait why did i i love it i love powerpoints uh so like i was saying we found these useful ways to design levels that supported ar

scenes that could have both interconnected moments where characters are talking to each other in groups and flowing in and out of conversations and also the private stuff that happens in these more private spaces we had found all these ways to design the levels to support that stuff which was great but we still had some elements of the level design that had to adapt to our ar scenes specifically so tacoma is a station where people live so for example there's a lot of tables and chairs and beds and various objects that the characters would naturally have to interact with while they're going about their daily lives um like wouldn't it be weird if like no one ever sat in a chair like that's just something like people do so you would expect it to see in these like daily life scenes right um so we're a pretty small studio with one one one full-time animator who animated like almost all of it with some help later on which is awesome but uh she did a great job and

but it meant that once the ar scenes were animated there couldn't be significant changes to like what the animations were because she was making a lot of animations and had to keep going so we could you know ship the game uh so we want to stay on schedule and if a character sits on a rock like clive does in the screenshot um that rock can't move a level designer put it there at one point noelle animated everything specifically to line up with it if we move it like she's not going to update the animation that's on us so we had a lot of animation dependencies like that that were in communication with the level design and that we as level designers had to work around and be aware of so here's an editor screenshot um if i was placing or moving objects or architecture i'd know that they were labeled as animation props animation dependent props so i would know not to move them and sometimes like sometimes you would also just like watch a scene to make sure

like you weren't placing something in like an awkward position that a a character was already moving through like if i wanted to do some set decoration i would have to make sure like don't put this like pile of garbage right where evie's like walking because why would she walk through that she would trip over it so we had to be very careful about stuff like that um because these ar scenes were being animated sort of simultaneously as we were like building these levels and placing a lot of stuff so it was all happening in parallel which was a lot it was intense uh so sometimes we actually took advantage of this effect like i was talking about not wanting to place the garbage where evie would clip through it while she's walking through it cause that would be weird sometimes we actually took advantage of that for narrative reasons which was fun because seeing a character clip through something could give the players some information about the time when that ar scene place that ar scene actually took place so basically it could help the player

understand the timeline of the narrative through observation of what objects were moved after the events in the ar scene occurred or perhaps objects that had not moved since the ar scene had occurred so for example in botany uh here we have sarah's leg clipping through the hose right at the bottom there where her foot is and later on in the scene you actually see andrew setting up that very hose as part of the scene he's sort of doing some maintenance to help the oxygen situation on the station so you can see him in this screenshot stepping over the same hose that she was formerly clipping through so while for the most part we tried to prevent erasing air scene characters from clipping through stuff in ugly ways uh sometimes it was fun to provide this kind of narrative detail where you could have the situation because if you think about it every level is like always in cutscene mode but a lot of the environment is like static like this hose so if you're

watching it carefully you can kind of catch these weird moments where it's like oh yeah this was moved during the scene and i can kind of see that and what what is the thing andrew was actually touching it's kind of fun to show the player like this is an actual object that this character really did touch at some point so the environment really needs to be in conversation with these animations in both time and space based on this example i just gave so while the level design of the floor plan was done before like the paper maps and stuff was done before these animations existed we were still able to have our air scenes be in narrative conversation with the environment using like set set decoration moments like that that were placed sort of later after the paper mapping and gray blocking was done so that was fun um so i want to go a little deeper into this time and space aspect of air scenes being in conversation with the environment um because one of our big challenges

was guiding players to these things called ar desktops which i have circled up there um error desktops are the characters computer desktops uh that they'll open from time to time at specific points in an er scene and when i say desktop i mean like literally like their computer with their emails and stuff on it like you can frob you can click whatever that thing the button in the air and it opens their desktop in your vision so given the spatial nature of ar scenes this means an ar desktop is only accessible to a player during a specific window of time during an ar scene so for example nat only opens that desktop for like 30 seconds a minute into the scene and then she'll close it after it being open for a couple seconds um because that's like the amount of time she takes in the scene to check her email or whatever she's doing um so early in development we found that

players were frequently missing out on these air desktops entirely during play tests like they would just like miss them not see them maybe be near it in a scene but be facing the wrong way for example and just like totally whiff on it um and then would move on and just have no idea that there was like all this extra story stuff for them to discover which is a problem for us um so era desktops often contain some like really great narrative beats so we wanted to fix this issue we had to learn how to guide the player to an error desktop at a specific time in a specific part of the level because otherwise they would just miss all the stuff so here is an image of one of the early era desktop button iterations in our editor and because the first approach we took to this was just like by relying on the buttons visuals and we were like maybe if we make it eye-catching enough like people will just see it easier which is like kind of funny that was our first approach now they think about it because it didn't work because the whole problem was like sometimes people would

just be in a different part of the area or like not looking at it at all so no matter how eye-catching it was they still weren't going to see it because they needed to be in a specific place at a specific time it wasn't an issue with the button being not obvious it was an issue with place and time so we realized this and it was a key realization because we looked at the error desktop timeline which i have like this is actually our first sketch of this idea which is funny from our whiteboard and that playing bar is like a rough sketch of our ar scene timeline that appears at the bottom of the screen that you're using to fast forward and rewind um and we realized that has a lot had a lot of free space on it that wasn't being used to convey any information um and we were like oh wait we could give the player like some kind of hint on the timeline as to when the desktops are appearing so we

are going to switch to the next slide here i've circled the aero desktop again and i've circled this time the pips which are what sort of came from that realization that they are seen timeline could be used to convey this information so we tutorialized this early on and stuff but it actually worked out really naturally most players pick up on it right away where it's like when you're scrubbing through the timeline those little bubbles will like inflate and turn to their associated character's color when the desktop is active in the world and then they'll like deflate when the desktop is closed um so players can kind of use that to be like okay how long's it open for what character should i be looking for we didn't just rely on color we have the characters um job icon on each one also to like double up on that information um and so yeah that was a really great solution to be like here is a way to give the player a hint but not actually like give them like an in-world like quest

marker or whatever where they're like looking for it in the level like they just have the information of the timing and then they do the exploration themselves to find it so that worked really well another challenge we faced early on in tacoma's development that's related to the ar scenes is that we wanted ar scenes and desktops to be tied to progression in some way but we didn't know what that would be at first so we wanted to do we tried some like traditional ah no i hate it sorry we wanted to do some traditional progression at first involving keys and locks because that's kind of what we were used to so we tried that first and then through play tests we quickly learned that people would find keys in desktops we put we put them in a lot of desktops early on and once players learned about that their whole goal was just to look for desktops and they would just fast forward through the whole scene not listen to anything

and just look for desktops because they were like the keys are all here if i just find all the keys i can progress through the video game and that makes me feel good and we were like that's great but we want you to pay attention to our ar scenes so we have to change how we think about progression to help the player actually like want to engage with ar scenes in the story while also feeling like they're making forward progress by doing that um so basically i pitched this thing called the sync device which is this is what it actually looks like in the game and i was like so these ar scenes narratively are this data that's been left over on the station like this this kind of like security camera-esque footage of what these people were doing there and we also have this concept of like going to retrieve this ai which would probably involve some sort of download so i was like what if in each wing of the station you

basically have like a usb drive type thing that you're plugging into that wing and like syncing with the data there as part of your mission to retrieve the ai like you have to sync with each wing in order to access the ai in the end and this i pitched as sort of like a a progression system based on something that we understand from our real lives so like people playing video games are probably familiar with downloading stuff or using a usb drive so they know when they see this plug into the wall and it starts ticking away with its little percentage meter players actually we didn't we thought this wasn't going to work but then it did players mostly are like yeah i see that download going i'm gonna go do other stuff because downloads are slow and we make it go slow on purpose and don't give the player a clear understanding of what exactly is like making it tick and they just assume it's gonna go and they can come back later and it'll all be fine and this actually works and we're really surprised and happy about

that because this sort of gave players the room and time to move through space and counter the story while they're sort of like okay that download's going so like i guess i'll just chill and learn what's going on with these characters um so the sync device in tacoma creates this illusion for the player um that the data is is being synced um and we don't have the sync device actually like if you stand in front of it and look at it it looks like it's ticking away steadily but it's actually under the hood not so the device's completion percentage which goes from zero to 100 is boosted every time you recover or completely finish an ar scene whenever you walk into a new room and there are a couple of other smaller factors that like make this the percentage go up a certain amount but we don't there's nothing pinging the player about this this is just happening purely under the hood so the boost based on player actions

help helps those who explore it because they're exploring and then they come back and they're like oh it's at 50 now that's cool i guess i'll go check out this other area and it'll probably be done by the time i'm back because that seemed to go pretty fast it was also really important to us for the sync device to be located in a physical space that was outside of the wing um we talked about having like a tracker so like the characters in the story amy who you're playing as also has her own ar desktop you can pull up at any time you get messages on it and stuff and we talked about having like a tracker that you could check on your desktop and we did not do that because we knew that at that point like it's interest it's interesting to make a progression system that is explicitly about being opaque and and not knowable to the player um and so if you could be like oh it's at 10 and then i recovered this scene and now it's at 20

oh i see so i should just go game this and find all the stuff to then again you would it would be like when people were just going into desktops to try to find key codes you would say okay so so now how do i touch everything to get this filled up as fast as possible whereas when you actually have to like walk down a hall to a space and start exploring it and then disengage from that to go back and check on what the progress is it both makes it less clear you know how to game it at all and it puts it out of your head until you're like i think i'm done here maybe i should go check that thing and and then it gives you something to to do to progress of your own volition without there being like achievement unlocked achievement unlocked just like popping up on your desktop as you play right and our sync device isn't very strict about this stuff so for most players just by like going to each area like it'll just go to under 100 without too much effort so we wanted to make it sort of also easy to do right while still requiring some amount of like exploration and

engagement with um air scenes um so yeah it was sort of designed to support these non-linear levels because the sync device doesn't really care what order you're encountering things in it just says engage with the environment i'll take care of the rest so that worked out really well for us um there's a sync device again and now we have some hilarious horrible art from me that i drew of it when i was pitching it uh so there's some fun stuff to show you sort of where we started with this we wanted to like we like did think about how how would amy carry this around that was kind of fun and then kate our environment artist drew lovely concept art of it which is great um so that's kind of what our initial design was for it so yeah that's sort of my end of things and i'm going to pass it back to steve and he's going to talk more about stuff thanks see doing a pair talk is great you get

so much more applause wow um so you know i think what nino was talking about is um you know from our perspective all of the elements of what we were doing um were really in conversation with each other um you know the spaces are connected to the fictional function of the station the ar scenes are connected to the shape of the spaces the player's hud markers and the design of the progression mechanic are connected to how the player engages with the ar scenes and how they direct themselves through each area and all these elements support each other and they loop back into each other right like the the the space the terms of the progression and the progression tells you something about the setting and the fiction and what your job is and that anyway so you know they this is none of them stand on their own they're all reinforcing each other um so from a creative direction and player experience standpoint uh that's really good uh you know we're talking about unifying all these disparate elements and making them feel like they are a cohesive whole

and you know this level of deep integration between all the parts of the story and the environment and the mechanics it also presents a lot of its own production challenges so let's talk about the writing and recording and the implementation of an ar scene um something that was uh interesting i think unique about how we how we worked is um we had to have our gray block layouts of the station locked down before i could write the story because the playable spaces in the station determined what would happen in each scene the the scenes are about these characters moving through these spaces so i needed to know the kitchen's over here in the conference room is over here and the dining table is here and the pool table is there before i even wrote a scene about how nat was going to call down to sarah at the pool table at this point in the story pardon me and i don't think that there

are a lot of games where the level layout being locked is a dependency on starting to write the story of the game um but that meant that we had to iterate early and feel confident in okay like these spaces make sense for the function of the space and for like a an explorable space fictionally and you know from the the player's perspective and feel confident in that because you know like like nina was saying about the animations having a dependency on just a prop being placed somewhere the story had a dependency on the level design not changing very much um so up on screen um is the the uh the the paper map that nina showed earlier the kind of final paper map of this section of the admin wing and then next to that uh is a couple of pages from my notebook writing the game i always write all my stuff longhand in notebooks and it can be really useful to you know

see revisions and see how things are connected you can see a little bit of highlighter here writing this game was um there's a lot of stuff that we did on this game that we really didn't have anything any examples to look at for the most part so we just kind of had to figure it out so in tacoma there you know the game is about there being these sections of scenes that branch off and happen in parallel and then come back together so there were parts where i would be writing a scene and then highlight like the blue part is this sub scene the green part is this sub scene they're happening at the same time and then they come back together um so the level design and the writing were really of a piece um and you know at that point yeah we kind of had to just like stick with it we had to we had to go from there five minutes well i'll be goddamned all right um so because as noted

we had to lay out the space write the stuff and then we had to go record it and we recorded um tacoma ensemble style there's mike cerrix i see him right there our voice director um mike cervix is here um we felt it was really important because all of these scenes were about characters being in a room together and having discussions and being face to face and having these intense moments for us to record the game on a sound stage with all the actors present performing the scenes together which was which was fantastic the the actors um were really excited about it because in games mostly it's you know one person in a booth and another person in a booth and then being able to perform this basically like as a stage play um was was you know them being able like react to each other's energy in in the scene and everything was really exciting um but it also meant that we laid out the levels wrote the story flew down to la got six actors together for two days and recorded the entire game and we weren't going to do that twice so that had to stick as well so we had like for people who have shipped larger

games voice recording on tacoma was locked 14 months before we shipped um because our animator noel had to animate the game based on those reads so like if we we could have done pickups you know like small pickups but we couldn't be like you know what this scene should we should do this should be something different here because then you know noel had to wait until we had the final recordings to even really start you know she could block out these scenes but to actually animate them again another dependency on um this whole stack of everything being related to each other which is you know part of what nina pointed out but something that we were really happy to be able to have is that because like nina was saying that we're basically in cut scene mode all the time it meant that the the animations could be very conversant with the environment so we did a lot of the characters touching the world so in all of these you know screenshots you can see

andrew's leaning on a rail and evie runs her hand along this railing as she goes up the ramp and nat is sitting on the edge of a thing and you know that's at that point another place where the gray block influenced the story and then the story influenced the animations and then the animations were touching a railing so if that railing was for instance grey blocked in when the animation was made and then we put the mesh in the mesh had to match the gray block down to like the centimeter or else the hand would now be floating or clipping or blah blah and you know that that went all the way down to things like placing this mug so at some point we were like we're gonna have a jazz mug in the game because we had a jazz mug and gone home and that's important to us so um you know uh noel our animator was like okay so this mug is here so how about uh clive is like carrying this mug and drinking from it during the first part of the scene and he puts it there during the scene and so we're like cool that makes sense and then at some point we were play testing we noticed that clive puts the mug like he like puts the mug

down and his hand just goes through it and it's like there's no way for us to move the object in such a way that he wouldn't just just do that just knock that mug over um and so so that was one of those points where it's like we do need to update update that animation because it's no way that it makes sense for this mug to be here otherwise but you know like um so um you know it's something like this where it gets the screenshots a little dark but we had a bunch of angled walls in the game and you know those were grey blocked in and and then it was animated for ev to be leaning on this wall but when the wall mesh came in if the angle was a few degrees different so those those are these like little things where we intended to build an experience where everything was talking to everything else but that meant that everything was talking to everything else the geo talk to the writing talk to the recording talk to the animation which then once the animation was in it talked back to the geo because the geo had to like you know re make sure that it was consistent with the animation once

it existed um so when you're starting out you should try to fully consider all of the dependencies that are involved in creating a unified design as early as you can not because you shouldn't do it because you should know what you're signing up for and as i noticed as i noted we didn't have a lot of precedence for what we were doing so we discovered a lot of this along the way but i think there is valuable space for people to work in that i think is easy not to do which is the mental projection of if i do this what are the five steps later and what's this going to touch it's easy to be like if we do this it'll be cool or if we do this it'll touch this one thing but really kind of trying to do the visualization exercise of like what is this going to lead all the way to um i think it'd be valuable even just so that you can expect those things before they happen instead of stumbling upon them when they do um one minute that's good because i'm on my

last idea so we're talking about doing a lot of stuff that hasn't been done before we're talking about doing you know getting away from a traditional lock and key progression but that said we also did rely on those traditional hierarchy of information or hierarchy of value approaches in supplemental information so the central space is very interconnected they're based on these ar scenes but we also wanted to encourage and reward players to fully explore all of our spaces including like the side spaces that nina was talking about and that totally goes back to stuff that we've done before we have like keypad doors that lock off optional areas and it encourages the player to say how am i going to find this keypad and then when you get in there we have to reward the player and in a more traditional game that might be like some rare ammo or upgrade points or something in our case it was just extra information that tells you more about the story because you did the work to find it but it isn't

central to the player's experience but that uses a very traditional approach to hiding keys making blocks be clear rewarding you for doing the traditional level stuff to get to get in there but we have to think of it in terms of again the writing goes back to like what part of the story are we going to put in here that it's okay for the player to miss but that they'll actually be excited about finding um so you know for whatever you're working on stuff that is traditional i think is worth like relying on as appropriate um even if it's not like the new fresh exciting thing thank you very much bernina steve tacoma is going to be in the igf pavilion if you want to play it