360 Approach for Open World Mission Design in Far Cry and Assassin's Creed

Hello. Hello. All right. Hello everyone. Welcome to to my talk. A little bit of boiler plate stuff to get out of the way. So, turn off your cell phones, obviously. You guys all got emails for the survey as you checked in. So, don't forget to fill those out. They're important. And also during the talk, if you guys want to stay for the the portfolio review after, there's a blue shirt dude in the back called Grant. He's holding his hands up now. He has a USB key. So, you can just drop your portfolio onto there and then we'll review that stuff over the lunch hour. Cool. All right. So, with no further ado, let's get it a little Yay. All right. So, to start off, I want to get a couple things out of the way. I wanted to set expectations for this talk just so you guys know what you're getting into. Um So, first of all, what this talk is going to be about. We're going to be talking about linear progression versus

open world progression. So, as you design your mission, it's going to be about a visualization tool. So, a method to look at your data, to look at the content that you're creating, to evaluate if you're uh spacing things out properly. You'll understand when we go through it. Um it's also a starting point. This is this is not going to be an end all solution. It's going to be how you start your design and start looking at how you're going to lay things out. Um and it's also like for a lot of you that are already doing this stuff, it's it's just framing things you're probably already doing, but didn't necessarily put the vocabulary to it. On the flip side, just so you don't come out of here disappointed that you didn't get the information you were looking for. Um we're not going to be talking about block out. So, like how to create your 3D setup at first. This is something another talk should be about. Um we're not going to be talking about mission mission scripting either. So, how to set up like the perfect mission like more

like technically. Um this isn't going to be about it. Um and most importantly, this isn't your holy grail. So, don't see this. Don't expect to be like a panacea like that this if I do this, my entire mission design problem is going to be solved. This is just one part of the process. All right. Now, let's get on to it. So, I want to do a little bit of personal background. Um it's helpful to understand uh the the the flow of of the conversation. Um So, this used to be me. I used to work in waste water management. Um so, technically, I used to shovel for a living. Um so, it wasn't the most glorious career. I was very well paid, but at a certain point after an unfortunate event, I decided I can't really do this anymore and decided to go back to what I loved when I was a kid, which was designing games. So, I went back to school, got a couple training programs in game design and level design. I specifically focused on level design and eventually got a really

nice job at Ubisoft and I've been there now for 15 years. Specifically Montreal. Um these are the games I worked on. So, I started my career on Prince of Persia: Sands of Time. This was a dream job for me. I played the old Jordan Mechner games when I was a kid and when I got the opportunity to work on the like new generation of it back in the PS2 days, I was like, yes, I'm jumping the ship. I'm going there. And then followed through with Prince of Persia: Warrior Within. I worked on Splinter Cell: Chaos Theory afterwards and then had a short stint on Shaun White Snowboarding. And for the most part, I would say like more than three quarters of my career has been on Assassin's Creed games. Just focusing on open world design in those environments. Specifically mission design. And I'm currently working on a project I'm not allowed to talk about, but maybe in the next couple years in GDC, I'll be here. Uh what I want to focus on today though were these three games. So, Prince of Persia, Splinter Cell and Assassin's to sort of explain the progression between all of those three.

Um do do do do All right. So, starting with Prince of Persia. So, for anybody who's not familiar with it. So, this was my first AAA game experience. I'd never I'd worked in MMOs before, but like really shitty little things that nobody's heard about. This was my big thing. So, I got to Ubi working on this. This was like So, it's it's a standard old school linear progression game. So, it's it's a platformer. Um for anybody who's done like classical level design, it's just going through a tube and setting beats, right? Um Our inspirations on the nav side were Ico. So, Ico was like a beloved project on for us. Combat wise, we're looking at Mark of Kri. Uh but game structure wise, it was much closer to the Jak and Daxter games. Like that that was our inspiration as far as level design stuff goes. What that means is there's not really options in the path of the player. Like once he gets into your level, he goes through a like what you orchestrated as

a level designer. So, you're really controlling his experience on a second to second basis. It's a it's really directed content. And as I said, like the player is like whatever you put forth, this is what the player is going to go through. Now, afterwards, I jumped onto Chaos Theory. Now, this is like it was the third in its series. It's it was a little bit different for me. It was a multi-path stealth game. Now, what I what I mean by that is is like so, when you were going through the game, the different environments were more like you had a single objective in any given room and you had more options to get through. So, we would just give players option a whole bunch of ingredients and tools that he could use and he would make his own way through it. Like we would have um profiles like clear profiles. So, like the more aggressive player or the more stealthy player and give them tools that complied to that. And he found his own way through that environment. But, it was still limited in size. It was still in closed spaces or very small outside like courtyards or

things like that. So, it was fairly focused. So, you're still controlling the player's approach, but instead of being on a second to second basis, maybe more on a minute to minute basis. Um and what that means is you can't really you can't rely as much on the the pacing or controlling your player as much, right? You're you're you're guiding him, but a little bit more loosely than you would in a straight linear game. And that led me afterwards to Assassin's Creed. Now, uh most of the team that worked on this, we were a lot of people that just finished Sands of Time and then jumped onto it for the conception phase. As I said, I went onto Chaos and then came back with the my original team afterwards. Um So, what we started working on wasn't what you guys saw in the end. Um initially, we were a lot more inspired by the POP game that we were working on before. So, the the the the first version we had was more of an epic

linear action adventure game. Um It was meant to be a linear platformer on steroids with multi-path. That's that that was the approach. So, a lot of the Splinter Cell stuff that I talked about like was there in our design. So, the different bubbles that you go to into would have a little bit more multi-path, but you were still going like from tunnel hub, tunnel hub, tunnel hub. Um And the scope was humongous on that thing. Like cuz the for the first year to we didn't have an engine. We didn't have Anvil. So, a lot of it was paper design. As you can imagine, you just go crazy, right? Like it's just free imagination. You don't have constraints. We didn't even we didn't know what the console specs were for the 360 or the PS3 or PS2 PS3. Um so, we just went out. Like we just drew a whole bunch of things on the board, wrote a whole bunch of scopes. Our original version had 72 missions in it. It was just We had like one of the missions we had was like in the Krak des Chevaliers. Like this big

uh Templar citadel in the middle of the desert. We had like this dynamic light towers that would invert as you played through it. So, like this big thing that would then go down like a funnel and then go back up. We even had like this giant snake ride. So, in the sewers under Acre, it's like a whole bunch Corey May or writers just went crazy when he wrote this Um so, a lot of it was cool, right? Like we were super inspired by it, but then reality checked in and we got the the first version of Anvil where like, oh, we can't do like a third of this. We're like, okay, we'll rescope and then eventually we realize we can't do two thirds of this. Um And we're like, okay, we're sort of screwed. We need to re uh we need to look again at what we designed and focus a little bit more our our design. Um So, that led us to the what you guys eventually saw. Um by looking at what we already had, we realized that navigation was our fun breakthrough. Like that was what what

was cool about playing the game. Um The other aspect of it was the crowds. Like it was the first time as far as I remember where we could get a lot of NPCs on screen and they were actually game playing ingredients. People you could play around with. They had reactions. And so, those two factors combined, navigation and crowds, were what we decided to focus on for what eventually became the game. Um and So, that like led us to like a densely populated, tight environment and we decided, okay, well, so cities cities is the way to go. This is how we're going to use these two features as much as possible. Um and that led us to the open world 360 degree assassination profile. So, and we we decided to focus on assassination cuz I mean, we were assassins after all. So, that that was an easy decision to make. Player free Sorry. So, what that meant though is with this approach, it meant that if I was going to open world, I could

not control my player as much as what I was used to in the past in the in the past. So, I had to get used to that idea again and recheck, okay, what does that mean for my progression? Um and in a sense like we were making our first open world game at Ubisoft. So, like there was a lot of questions that that we needed to answer before we can move ahead. So, I was left with this question. How do I manage my progression? Um now in the past, I was used to once again a more linear approach. Uh and going into this big open world, um I needed to check what what what that meant for it. Sorry. So, my first inclination because of my past experience was to say, "Okay, I'm going to I'm going to get I'm going to design player paths, right? I'm going to design a stealth approach. I'm going to design an acrobatic approach. I'm going to design a fight approach into each of my setups." Um that was a bad idea.

Uh cuz I realized with the open world, I there was so much landscape or so much areas to cover that I couldn't handle it all. There was just too much data and So, I would Sorry. Um I have too many notes. Okay, and I see Yeah, as I said, I I realized that there was no way that I was actually going to cover all this stuff. So, I need to reassess once again what it meant to do open world progression. Um So, I started to thinking back to the my pop experience. So, if you consider the path of the player takes, most of the time like when we when we would draw out the progression in the map or the the player path, like we all do like a golden path, right? It's the the where the player comes into the area and where he's going to go. This is one of my the examples from Sands of Time. I was the water reservoir. Um so, the the like the initial the the the the the start of

that spline is like where he comes in and the end arrow is where he gets out of it. And I can sort of define that path through like one simple spline. There's very little branching other than for collectibles in there and there's no expression on the player's standpoint. So, it really comes down to one simple line. If I take the open world setup, usually what will happen is you'll set where your mission location is or like your target or your objective and then you draw this like circle or blobby area around and that defines the general area that your mission is going to take place in. Um And if we apply the previous knowledge, then you can sort of represent the like the shortest path your objective as the radius of your circle or like a single line that goes from the outer edge of your uh of your area down to the creamy center where your main objective is going to be. Um but really this is what your player is going to be doing, right? He There's no way he's going to be going straight from the

outside to the inside cuz he's going to be exploring. He's going to get familiar with the ingredients you have in there. He's going to move left and right. He's going to go backwards. He's going to like circle around your objective. At some point he's going to commit and go down to the center. Um and that became super interesting cuz I was like, "Okay, like the the area is where things take place. So, how do I how do I uh make sure that there's the player can feel a progression from that outer edge to the inner edge? Um with that that like the the circumventing that he does, it it it's not a direct path. So, it became a little bit harder. And this here, I'll give an example like an actual in-game example of what that means, right? So, this is uh an example from AC1. Um this is the Sibrand assassination in the port of Acre. Um so, we we'd already defined where the assassinations were going to take place at the beginning of the project. So, I already knew like this guy, he's going to be in the port. Like that that that already knew. We already done like a sort of blockout for the area. That was mainly inspired by architecture

uh and like just historical references that we'd found. So, we just mocked up a thing, made sure it looked like a a port with like boats and in there. And like this is what it gave. Um And now in there, I placed my target. So, it's that little crosshair uh at the bottom right on the boat. Uh so, that that I knew or I I had established that this is where I want my my final moment, the crescendo to happen because it was in the middle of the water. The character didn't have any swimming abilities back then. He would automatically die. So, it was my lava pit. Um so, it just made sense to have everything progress towards there. Um and it was a nice like amphitheater setup if I did that, right? Like from the outer edge of my city and I'm leading towards everything sort of naturally focused around there. So, it just made sense. Um afterwards, what I did is I defined a zone around it. So, that as I said earlier on, that's my mission area. So, that's that dotted line around the whole thing. Um so, in a sense, the port The port is where most of my gameplay is going to be. Anything that's outside of there is an approach. I'm not going to

control it. I'm going to use the systems of the game in there, but I'm not going to put anything really in place to sort of complement my my mission right away. I'm going to focus mainly on the port here. And as I said earlier on, the shortest path is from the entrance of the port directly to the boat, right? Like that's going to be like if you want to do the mission inside of 5 minutes or less, that's probably their best path. Um but really most players they approach from anywhere in there. And so, you're going to get these player stories are going to be different. Everybody's going to have their own way. And in a way you sort of have embraced this. This is like you're making an open world game and this is what you want. You want everybody to have their own story. You want everybody to um tell you how they did it and then the next day you do it in a completely different way and that's the upside of doing an open world uh setup. Um Sorry. So, with this in mind, um that led us to

take it on a more systemic approach. Because like the whole area could be used, there's no way I could control the whole thing. So, using systemic ingredients where I we would prepare a whole bunch of either ingredients or ingredient setups, uh we would create like a list an ingredient list that we can just distribute through there and the player that could then interact with all these toys. So, similar to what we would do in a small multi-path room in Splinter Cell, we would do this on a wide scale in a large area. The one thing we're scared about though is as opposed to Splinter Cell where everything is focused in one small room, being a large area, you can lose the focus. You can lose the the the intent uh that you're trying to go for. And it'll just in the end there's a risk that it just might seem random. So, once again, I need to think, "Okay, well, how do I sort of redo that sort of progression curve that you could have in a linear game in this setup?" And so, that brought me to segmenting my uh my areas.

So, if I was to divide my mission area into multiple concentric zones, then I could sort of have a like a similitude of what I was doing in the open world progression cuz normally they are Sorry, in the linear progression where like you'll have beats for each given room as you progress towards your objective, I could do that more in an area using this approach. Now, I'll give you an actual example of this. So, I'm going to use Assassin's Creed again. We're going to use the same example from what you saw earlier on, but we're going to go a little bit more in depth. Um so, first you need a rough layout. Like if you don't have a rough layout, you're going on a blank slate. Like you're just going to hit your head against the wall. At least try and block out like a general idea of where your your your mission takes place. Just like you could I've done versions of this that only has like five blocks in it and and and a ground. Um just to have something to work with. If you don't have anything, it's just it's going to be too long um and you won't get any good results. So,

just at least try and do your blockout first. Um and so, here this is this is like a much more complex blockout than what you need. This is one of my final versions, but I couldn't find the one of the older ones. Um but you could strip like probably half the content in there and still look like a port. Um Next step is identifying where your goals are going to be. Um so, in this case as we said earlier on, we had the dude on the boat. I I knew that's where it was going to be. Afterwards, you define your approach zone. So, we already talked about that one. So, just define your area. Um now the size of the approach is basically it's it's up to you. Like it's dependent on your game. I can't tell you the size of it cuz it's like the character movement, the speed that he goes at, um what type of objective is your mission uh or even the mission time. Like say it's something that you all have to agree upon on your project. Like do we want our missions to be 5 minutes, 10 minutes, 15 minutes or I don't know, an hour? And that defines the area that you're in. Um

And then from that point on, once you have a general area, then you want to segment it up just to create sections uh that progressively lead you towards your objective. Um now you might want to avoid creating too many of these uh areas because the more you have, the less contrasted you can make them uh be. So, the larger area or the the lesser you have, like the more distinct you can make them all be and you that that progression or the the passing from one area to another will be easier to sort of uh indicate or make the player feel. Um Sorry. Um now you can see here also that like I it's not concentric in this case as I said earlier on I I built an amphitheater setup so everything leads towards the boat. Um,

this just came about because of the environment I had. Everything like the water I can't navigate in so it just made sense to have sort of progress towards it. Um, in a case like let's say an outpost from Far Cry where it's in the middle of a field or a much more open area then you'll probably have concentric zones that like so a circle in the middle and then you approach from any given direction. So I guess like here what I want to say is take inspiration from your world as well as from your abilities uh and have fun with your with your zones like they they don't have to be like concentric geometric shapes like really take inspiration from the geometry already have and you can use that to sort of guide your player without once again trying to limit him too much. Um, and this is a like a good thing to know is like you're once again in open world like try and not limit the player's options try and give him as many options as humanly possible. Um, so that he like you can get more player stories basically out of it in the end. Um, Whoops, sorry. All right. Now the next step this might

seem a little bit like it cosmetic but I like to add color to to this. Um, this is mainly to improve readability of your your design document um cuz if you don't do it like ultimately in the end you're going to have a lot of information layered on this and if you don't do that color coding it just it'll make it much harder to read. So I just like to have this to like when I can take a top view of my whole thing I can see the contrast a little bit better. All right. Now the next step which is the the I guess where we I actually start getting into gameplay stuff um ingredient distribution. Um, so I'll go into details about ingredients a little bit later on after this. I just want to give you guys like a really quick view of how the method works first. Um, So basically what you're doing here is you're laying down the different ingredients that are complementary to your mission within your different progression sectors.

Now these ingredients can be anything from existing gameplay objects that you already have in your in your object bank. They could be art assets um they could be sound events. They could be gameplay patterns that you want to use. Um, pretty much anything is okay as long as it is complementary to the theme that you're trying to represent or the type of progression and we'll go into the into details on that a little bit later on as well. Um, And here what you want to do is basically like the closer you get to your objective um well either the harder the the ingredients become or the stronger the emotional ingredients or like the strong the the the closer they are to the target emotion you want to get um depending on once again the theme that you're going for. Um, one thing to be careful of is don't increase the number of objects in each and every zone. Think about density not about number of objects because naturally as you get closer the zones

become smaller and smaller. So if you have more objects you'll just like have way too many ingredients in there and like you won't it'll be chaotic in the end. So just be weary of that notion. All right and really this is where this is where it gets important is this is all document this is all theory but you're making a game so everything has to be in engine right? Like as soon as you can you want to drop this theory into your engine and start playing around with it. Like this is where Sorry for the pun. This is where the magic happens. Um, but like this is where the game is right? Um, So usually what I would do is I would take that theory so the the the stuff I would set up and then drop it directly into my world layout and then adapt as I go. Um, now the first thing you're going to notice is you were using a 2D tool. So the 3D component to your game is just completely busted. You won't like you'll you'll you'll distribute stuff evenly because

of your top-down approach um and like verticality won't be considered. Um, you'll notice that distances are off so then you start distancing objects from one another um especially when it comes to mobile ingredients like enemies cuz you'll get aggro. So if you aggro an enemy and you bring him to another enemy then like the difficulty just increases exponentially. Um, so this is stuff like you'll notice right away as soon as you start dropping stuff in. So like space it out as soon as you get in the engine. Like don't spend too much time on documentation just do it directly in the engine. Um, This is the the the like the actual final layout that we ended up with. This is the mission uh this is the map that was in game. So a lot of the the geometry I just readjusted afterwards. Uh the guard placement also I removed guys from my initial plan because it was hard to understand the the contrast between them. Elevation so like that this is where we realized that like enemies cannot see beyond a certain threshold too much so like if you're doing it was cool if

you're trying to do a stealth approach like getting people like if you're if you want to navigate on top that and enemies on the floor cannot see you um but it also meant that if you were planning something that had a lot of verticality like your intentions might not work in the end because enemies couldn't see you. Um, so it's just stuff once again directly in the engine you're going to notice this. Now I talked about it a little bit earlier on but how do you choose your ingredients? Um, So in your game you probably already have an existing list of ingredients hopefully or preparing the process of of designing these things. Um, so you probably have like a really wide selection of of there's just a lot of things that that you can choose from as from the art assets from the sound cues to any like small little scripted events. Like there's a lot of objects that are in front of you and so how do you choose which ones best complement your mission?

So what I like to do first is come to a common understanding on your project of what you want to challenge the player on. Like what's your I'm calling it a dimension here but really what is your uh what is the value that you want to see a progression on during your entire game? Um, more often than not we choose difficulty. Uh difficulty is the one that we often refer to um and that's a great one. Uh I I I highly suggest using it. Um, but there's other ones that you can use as well. Um, the mission phases uh what that means is like for a lot of military games you have the approach zone the infiltration zone and then the encounter zone. So you can use that as your progression um parameter. Gameplay focus is another one so I'll give examples for from Far Cry a little bit later on but like if you choose a gameplay object or a gameplay feature in the game and design an entire mission around there um

that could be another dimension that you're choosing to focus on. Um, emotional theme I'll talk a lot about these as well. This is one thing that we do a lot of on the well especially on the older ACs where we choose I want to do vertigo or I want to do uh fear I want to do um risk uh these these are it's it's a it's and this is another type of approach. And finally here there's probably a lot there's a lot more that you guys can come up with um but one I've had a little bit of experience working with was the narrative acts. Um, so this is like a lot of our background was work like just writing. We had a lot of writers on our team so we used a three act structure to to sort of inspire ourselves on mission progression. So we would design an introduction zone so how you get where you get familiar with your concept then the catalyst being the event or the cinematic that led you into the development phase of of your mission. So where uh the story sort of progress so the area where that happened and then the conclusion phase with the catalyst in between which was once again another

cinematic or something that led you into the final epic battle. Um, so it was neat intellectually but where it fell apart is all those choke points in between the catalyst where we were sort of breaking the uh the appeal that open world had because if everything is open world but then you're choke pointing your player through a specific point then you just lost everything that you've worked for. Um, so that one I would actually if you can do it where you're not choke pointing your player it's way better so think about it but if you have to choke point your player into specific moments then you're really throwing the the open world approach out the window so probably not the best thing to do. Now in the previous example I showed you before uh we used two dimensions. So like this is also a possibility. It becomes harder to sort of visualize your data um but it is something that you can do. Uh we were using here uh so difficulty progression as well as emotional progression. So in this case I was using exposure as my my emotion. So the

further you went out to the water the less um objects you had to sort of block line of sight with enemies so and you were on water. So the the closer you got to the enemy the more exposed you were to the different enemies um and also the risk of falling in the water and whatnot. So like it's it's just one way to progress things. Um, now once you have your dimension uh isolated choose a theme. So this is the flavor. This is uh so if we were talking about emotions earlier on well okay I would like we generally on the project decide emotions are what we're going to bank on then is this mission to be about vertigo or is this mission going to be about love? Is this mission going to be about like eating? I don't know. Um, This is just like the subset right? It's it's the flavor that you want to choose for your given mission. Now, the example I gave here is for gameplay focused on on AC, like the main pillars were acrobatics, stealth, or fight. So, we could decide this specific mission, we're going to make it a stealth mission. So, this this is what

that what I meant by this. And finally, listing your appropriate ingredients. So, this this is taking all the ingredients you already have in your bank and then picking and choosing the ones that are that that fit the theme that you're going for. As I said earlier on, these can be anything from like actual game ingredients you already have in your bank. They could be scripted events, so they're like just a conversation between two people, explosions, things falling apart. They could be sound cues, so just hearing like a bird chirp, or like people screaming out in the distance, the wind, like that type of stuff. Art assets, so like BioShock for instance, like all of the graffiti that they would have in there, like the the environmental storytelling setups that you could do. That fits pretty well in the art asset section. Gameplay patterns, so this is more of an amalgamation of multiple gameplay

objects together. So, when you see like when I was talking about exposure in the previous mission, we had a lot of moments where we would put what we call troublemakers, so they were the like the crazy dudes that would just like push you. Put those next to really narrow platforms. So, that was for me a gameplay pattern where putting guys near narrow platforms where I can fall, I can re-instance that a couple places, and that became a gameplay pattern. And finally, this like also list things that don't exist. Like you don't have to go through the list of things that you that are already there. If you find that you're missing stuff in your list, like on any of these standpoints, you can propose things, right? You can come up to any manager or content organizer and say, "You know what? I'm missing ingredients for this type of emotion. I'm missing the ingredients to challenge the player on this type of gameplay focus." So, just come out with your own proposals, prototype them if you can, and then like you can help develop that

list of ingredients that can then be shared by other people. Um Now, the more variety you have in this, the better off you'll be. Like the more objects you have, the better off you'll be cuz especially when you're dealing with an open environment, if you're reusing the same object over and over and over and over again, that's cool in the sense that the player is going to like recognize it and be able to deal with it, but you end up having the like the Flintstones syndrome where like it's just the same thing repeating over, and you notice the pattern and it becomes less compelling. So, the more deep your list goes, I think the more flexibility you guys will have in the end. One thing you might want to do is also create subcategories in there. So, if you choose a given like emotional theme or whatever, choose things that will complement the progression in there. So, things that are like weaker in emotional state to stronger in emotional state, or something that like flips to something that is like a different dimension of those over

time. So, that you then once you start applying it in your in your 2D method, you can actually segment it per world area. Um And also like an a neat thing that you can do with this is this is where as I said earlier on, you can notice, "Oh, this gameplay thing is completely understaffed in ingredients, or I don't have enough objects to represent this type of gameplay focus." Uh And on the flip side, you can actually notice the stuff that you're using way too much. So, it's a way to sort of do an evaluation on the objects you have in your bank and and get a an idea of like, "Do we have enough? Do we not have enough?" So, it's just another way to evaluate your own content. Now, here's an example I wanted to give for emotional states. So, I mentioned vertigo earlier on. So, this is an example from Assassin's Creed II. It's the San Gimignano assassination in Tuscany. So, we we tagged this mission as

a vertical mission. So, going through the list of of objects we had, we isolated a couple ones that that seemed to complement that a little bit better. And so, we ended up with like the swing pole. So, the idea that you you had like these poles that only your hands would touch, so that meant your feet were not touching the ground. So, all of a sudden like you lost a lot of your security. So, it played a little bit more with with vertigo. Corner swing, which were those objects that just like turned a corner 90 degrees. And so, that one you're putting your your uh your livelihood in the hands of the designer, right? Because you didn't know where that was going. Just hoped that it led to something cuz there's a way if the designer knew his job, there's a real risk that you could jump onto those and then fall to your death afterwards. So, that that was another one that helped us a lot. Archers, so as you climb up, having archers on opposing towers that could spot you, put you super exposed, and once again, you became very conscious of the the lack of floor

underneath you. Um tightropes, so this is a an easy one, just having these really exposed tightropes between towers that you can just plunge underneath. Um and once again, exposed to those archers was really good combination. High buildings, so as you progress, and this is a feature of the San Gimignano towers, like these are like really tall towers that like peek out from the really low horizon on all those buildings. So, that already just in its architecture helped us out. Um on a less gameplay side, we start we put in flying birds. So, like the higher up you went, you ended up like, "Oh, wow, I'm climbing amongst a whole bunch of like flocks of birds." And so, you became very conscious of how high you were cuz normally you wouldn't see those. They would be so high that they were just like background elements. But when you saw them next to you, like, "Oh I'm really high now." Um and finally, the wind ambience. So, like going from a more uh city atmosphere when we're at the bottom, and then progressively switching that to a wind atmosphere once we got high up. Same thing, made you really conscious of

the the risk that you were now taking. Um now, here's an example from a game I never worked on. Um but like I'm sure you're all familiar with the Fort Frolic mission from BioShock 1. I was talking with the mentors later. They they gave me a a little bit of insight on on the mission, and I found it really interesting and actually very appropriate. Um So, the Fort Frolic mission came to in in a moment in the game where you were getting pretty confident with your skills as well as the like the eerie nature of Rapture. And so, it was put in here to sort of like destabilize you from that like that the position of safety you were sort of getting. Um So, from my understanding, the intent from Jordan, the level designer on the mission, was to destabilize the player. And to do so, he went through in in a way like these enemies did not exist, right? Because they're not used anywhere else in the in the game. This was like a unique

place where the these are used. So, up until now, the player's been encountering dudes with guns. For the most part, it's just a whole bunch of dudes with guns. And here you start encountering enemies that can teleport. They mainly attack you in melee. Like the only other enemy that attacks you in melee was the the Big Daddy, right? And he was like a big thing, so you already associate to something that, "Oh man, like this is not going to be comfortable." The fact that they walked on walls like was not something you were used to either. You're used to aiming at eye level and just shooting heads, right? And because they could walk on any surface, that put you at ease. Like your aiming needed to adjust. Um and finally, they use smoke grenades as well. So, it just like and they could disappear on top of that. So, it just put you completely uh uh uneased. Um and finally, the graffiti once you like they used that in the rest of the in the game, but like it just helped to complement that really really well.

And finally, my this is my last example. I wanted to give an example from Far Cry 4. Now, once again, I didn't work on this, but I sat down with my colleagues from from Far Cry. So, Nicolas Duclos, their level design director, as well as Danny Pockett, their outpost designer. Like he's the guy who designed pretty much all the outposts in Far Cry 3 and Far Cry 4, so he knows what he's doing. Um So, here we have an outpost. The example I wanted to talk about is the Shanath Training Ground in Far Cry 4. So, this is the for anybody who's played, it's the dog kennel outpost. Um Now, their focus here, it was a gameplay focus. They decided to focus on dogs. Now, dogs were really different object from all the other enemies because well, they're animals. They're not human beings. They don't have weapons other than their teeth. And so, the the the the the specific gameplay nature of them were they're they're basically meat missiles, right?

They see you and they just run at you and chow down. And they don't have as much strategy as the other enemies have. Like they won't take cover. They really like if they see you, they jump at you and and try to bite you. Um So, to find the best environment to sort of suit this gameplay focus. Now, this is something the the Far Cry team worked a little bit different than what I was used to on AC. But well, what they'll do is they'll do Google Map scouting. So, when they know the general area that that their game takes place in, they go and take satellite footage of like different areas and try and find something that they find neat, like a a distinct feature, and try and to that in layout. So, here like they found a some farm setup I don't know exactly where this is, but somewhere. Um and so it took a screenshot and then okay, we're going to try and reproduce this. And then as what I was saying earlier on about like very simple block outs like this is a version of that, right? Like there's six blocks in here

plus one cliff difference. And you could start designing stuff with a block out that is this simple. Um so they reproduced that image that we just saw. Um And then all right, and then then they started applying mission content onto it. So, this setup is way simpler than what I showed you earlier on and that's intentional. This is just a to focus on something that's maybe easier to approach. Um You can already see it just from this layout. You can I think I think like you can sort of define how the areas are going to go. So, let's see if you guessed properly. So, when you start mocking it out, you can see like the initial of that center zone which is that big open area. Now that's really appropriate for dogs. It's a big wide open area. As soon as you get in there, they can spot there's nothing to really block them from approaching you so they can run directly at you. Um and that secondary zone around it is

basically the road side and so this has more line of sight blockers. Dogs will have a little bit more problems seeing you. However, if they do see you, they can go through that geometry and try and get at you which makes it a little bit harder for you to shoot that and we'll get to that a little bit later on. Um So, here you can see also like well, we're going to color it. There's less zones than the previous examples that I gave because it's an outpost. It's not like it's not a mission, right? The missions for ACU like they were meant to be 15 minutes long. Outposts in general you target about 5 minutes as a general rule. Um so like you you'll naturally have less zones in there. Um And so an interesting thing here is like only the red and yellow zones are actual gameplay zones in here. The green zone is what I was talking about a little bit earlier on. It's an approach zone. It's it's an area that has no direct challenge to the

player. It's more a get familiar with your environment zone. Um so I'll give a couple examples of that in the next slides. Um But as I said, it's really it's an area that you can use to your advantage to sort of get the player familiar with the general things he's going to encounter without necessarily challenging him. And you don't want him you don't want to put anything in there that's going to like alert his presence to the enemies unless like he's being stupid and throwing grenades everywhere in the middle of the forest and trying to kill trees for some reason. Um Like this is really just like get get the layout of the world. And so for anybody who's played any of the Far Cry games, um it's also where you usually get your your points of view. So, elevated areas that you can like just get higher ground and see what the hell's going on in safety for the most part. It's very little risk of exposure and this is where you will make your plan as a player. Um So, here like there's like basically three elevated zones around the the area

where the player can do that. Once again, it's an open world so you need to prepare for that, right? You need to prepare that the player's going to be approaching like here there's a weakness on the left side. If I'm approaching from from the west side here, there are no vista points in there, right? So, luckily there's three other ones so like as you approach you can see that there's higher levels and so the player can just circumvent and try and like go around the setup to eventually get through there if he wants to have more information or he could just go in guns blazing without the information needed. And that was intentional here actually. The main road leads into there so it was like the way to go the fight route in here. Um Now, once you have like that approach zone sort of set up, this is where now you now you can start distributing ingredients in there. So, we said it's dogs. This is the focus of this mission was dogs. So, the center of there was the the like the penultimate area. It's multiple objectives here. You have to kill everybody basically to win the challenge. So, the majority of the dogs will take

will be present in that little area. Fighting more than one dog is obviously harder. Um And they have little geometry to sort of block them from from seeing you. Now on the outside, a little bit less dogs. They're isolated one from the other. The two at the bottom there are actually in cages so they can't really get to you. You can shoot them through the cage or you're unlucky then you just broke the cage and they start running after you. But this is where you get familiar with the gameplay object and you're able to take it out by yourself without it warning anything else, hopefully. Now next step is putting in your guards. Here the designer put in only one single human guard in there and that was intentional once again. Like the focus is on dogs. You don't want to dilute their presence by adding too many enemies cuz at that point it becomes a game about guards that have dogs instead of being or instead of being a setup that is about dogs with

a couple guards around. Um so that was intentional here and he had this is the the hardest guard in this general setup. Not in the game, it's just a second level guard. Um but of the area this is going to be the hardest guard. So, it just makes him stand out from all the other ones and he acts as a handler. Now, he added the weakest guards in the game are these these three guys that he added. These they mainly act as warning bells, right? They like for the guards. Like if these guys detect you, they will alert the guards the dogs and then the dogs come running after you. And he placed them in areas really close to stealth entrances. Um so they're really easy to take these guys out. So, anybody who's done the job of properly scouting can take these guys out individually and then he's left to just dealing with the dogs. Uh and then afterwards he does the passive like the boiler plate stuff of the reinforcements, the the any flammable surfaces and the alarm bell.

Um So, a lot of this like you can imagine is reverse engineering of what some guy did, right? Like he didn't do this job. He did it instinctually. Most of like the the the the defining of the zones, he did that just by looking at the content. So, as I said the very beginning, this is probably stuff that a lot of you are are doing, but it's just by adding that visualization layer you can realize if you're doing your job correctly or not and if there's stuff that's missing. So, here by doing this this this exercise and I talking to the designer, we realized that that green that that green zone could have been used a little bit more than just a general approach zone. Like we could have dropped in empty cages to warn the the the player that hey, there might be dogs coming up or mangled beasts or civilians that had been mauled by a by the guards or like just those weaker guards again that are just like calling out and trying to to to find their lost dog or something like that, right? So, you're just warning the player that these ingredients are coming without once again creating any form of conflict or

any challenge area. And then you get into the engine and you adapt. You you need to actually build this thing, right? Um so as soon as the guy had planned it out, like they they spend about like a week less on this document. Same thing for Assassin's. Like we we don't spend that much time on the actual documentation phase. It's more to get like your brain juices flowing. Um so they dropped it in, started building. This is the final setup. This is what it looks like. Um Now one thing they noticed, I mentioned it earlier on, was that covers are really really really when it comes to dogs because the dogs they're lower, right? And the player as I said from the BioShock example, you're used to aiming like a head level. And so the dogs are so much lower so you're actually aiming down and that's really really hard for the player to do. Um If they're coming in a straight line, you have a chance. You can actually like track them and shoot them down, but if they're dodging and darting through geometry, that makes it super hard cuz like the it's just something that

enemies don't do normally. So, they ended up removing all the cover and then more focusing on like when you try and leave that central area, creating lanes that you can sort of dart into and then when the dog is coming at you, it's just a straight tunnel. So, it's actually just a shooting fish in a barrel almost. Um and once again, this is only stuff they could realize once they got into the engine. Um There we go. Now, what sucks about this method? Uh now I've been using this for 10 years now. Um I'm well aware that it's not perfect. Uh it really is as I said like a starting point. There's a whole bunch of things that like don't work so well about it. Um It's really dependent on your block out. As I said at the beginning like if you don't have a block out, you can't do much with this. You so you need to actually create content before you can start even thinking about applying this. Um same thing for the gameplay ingredients. Like if you don't have

a list of gameplay ingredients, you're a little bit screwed. Like you're going to get blank page syndrome. Um so I highly suggest getting some of that stuff out of the way before you start applying the method. Um Sorry. Um like any documentation, it gets quickly outdated. I'm sure you've all worked on documents that nobody reads or that like the if you look up your final documentation in comparison to the final game you created, like they don't match really. That's okay. Don't worry about that. That's actually a good thing. Um documentation is not meant to be kept. It's meant to be thrown away. It's mainly meant for your brain to get used to the idea and start playing around with Uh, as soon as you have that, get into the engine and start building stuff. So, this is actually it's Yes, it's a downside that it's quickly outdated, but it's not the end of the world. However, there are interesting things that can be done. Um, the Far Cry team actually developed a thing internally where they

they call it Atlas. It's It's a visualization tool that keeps in sync with the engine. So, anytime you move around enemies or uh, like any gameplay object, there's actually like a position that is known that is sent to the tool. So, you can do your prep work in that tool, go into the engine, move things around, and then as soon as you move it around, it's updated to the the the actual file. So, it's a way to sort of keep your document alive and and living, but once again, don't don't rely too much on it. In In-game is better. Um, And finally, oops, sorry. As I said earlier on, it's not 3D. That's That's the biggest problem with the tool. Uh, it's all 2D. So, for the most part, like on AC and on Far Cry, it's not the end of the world because yes, the game is 3D, but it's not really, right? In ratio, when you look at your world, like everything is so spread out that the height difference does not match really

the the spread that you have on things. So, the 2D method is not the end of the world. Like it it it actually it functioned in in the state. Um, where it doesn't work as well is if you have a lot of verticality. So, if you're like the the towers in Fallout, like all those those really uh, high sky rises that are networked in Fallout. Like you wouldn't be able to do this. Like you'd lose a lot of that verticality info in there. Um, the the high rise section in Sunset Overdrive, that was the same thing. Like everything is so high that you're like you end up actually looking at your level more like on a profile view than on a top view. Um, or like anything like in space, right? Like if you're using Star Citizen, where it's really 366 axis approach where you can approach it top, bottom, left, right, any direction, you can't apply this. This does not work for that. Um, so, looking to the future Look looking towards the future, I'd be curious about trying the 3D approach. So, using a 3D mock-up tool. So, either SketchUp or 3ds

Max. If you can do it in your engine like simply by um, putting like colored volumes in what not in there and creating really like a 3D sphere. Uh, as you get closer, if you Obviously, these are geometric shapes. Often times, when you look at triggers, you want to be able to sort of make composite shapes that like lead up towards it. But, in the future, I'd like to try this using this type of approach. Um, And in conclusion, these are the the the steps that we went through for for the approach. So, uh, starting with your top down, uh, identifying your zones in there, listing your ingredients that are complementary to your uh, mission design, distributing the ingredients, and then more most importantly, getting in the engine and actually building the thing. Thank you. That's it.

And now we'll take questions. Hi, thanks for the talk. Thank you. I really loved Prince of Persia, Assassin's Creed, Assassin's Creed 2, all those games. Uh, big fan. Um, I was wondering, you know, as a level designer, you're building gameplay on top of kind of block-out architecture. Um, and you kind of naturally drawn to the paths that that architecture elicits from you. And I was wondering, you know, like landmarks draw players towards them, bridges draw you underneath. And I was wondering if um, you had any tips or tricks for that kind of 3D architecture stage of the level design. Um, so on that standpoint, usually what I try and go through is creating uh, like a grid of landmarks uh, that are like the the the main navigation points between objects. So, like especially in things like in Assassin's Creed or like even uh, Fallout where you're recreating a real environment, right? There are landmarks that you sort of expect to see. But, if you're not doing that, a lot of the multiplayer games do it like

with landmarking uh, certain regions of it. And then everything that's in between is more like the lanes that go in between them. So, by creating those points of interest that you go from one point navigating to the other. So, if I'm any given point, I can see a landmark or there Okay, this is where I'm going to go. And then you end up like creating a grid, right? Uh, so when I was on Assassin's, the general rule we had is on any given landmark, if I'm if I'm near one, I can see at least two other landmarks from that position um, that are going to be like 100 m away. And that creates like my meta navigation met uh, method that I can just move from then one point to the other. That sounds great. Thank you. No problem. All right. And uh, my question is um, so kind of especially with Assassin's Creed where you kind of have these especially round environments usually in a city and you're usually kind of targeting one area of the city. Uh, and there's a lot of these kind of side objectives and side missions. How do you um,

how do you kind of keep that interesting for the player when they're kind of intended to go from point A to point B, um, but you're trying to keep them interested in moving around kind of doing a lot of um, what sometimes can seem very repetitive? Yeah. So, you Yep. I Sorry to cut you off for a second. We're going to do a few more questions uh, for Philip's talk. But, if you're here for the portfolio review and you have materials that you need to get on your USB drive, uh, Grant standing here behind the mic, he has a thumb drive that you can use to drop your files onto. We can get all that stuff prepared so we have minimum amount of bumps as we go into the lunch time review. Okay, no problem. Um, so basically what we do for that is So, you have your your your main line structure here, right? Like uh, going from one main mission to the next main mission to the next mission or even parts of a main mission to the other one. And those are going to be distributed throughout your city or your your general game space. Um, And so, good practice is actually have like a lot of distance between those so that the player is then navigating through your open world or your systemic

environment and then creating opportunities or or just distributing opportunities in there for him to sort of peek his curiosity, right? So, if I'm leaving here uh, to there, then putting like uh, like a pickpocket challenge or a couple treasure chests on the way. And that creates like uh, the first game I remember doing that is playing Crackdown, where I was going to a mission objective and seeing like those glowing orbs and then realizing 2 hours later that I was still on my way to that given uh, objective, right? So, it's it's good practice to do that. Like just space things out so you get you give a chance for the player to actually uh, experience your game world. Okay. Thank you. No problem. Hello. Hey. Uh, thanks for the talk. First off, I just had a quick question. Uh, what are your thoughts on, you know, as the port for an example, you'll maybe spend an hour or two in there outside of a mission. Uh, and that will have an established structure to it that the player will learn. What are your thoughts on kind of subverting that established structure for the sake of a narrative? You know, if you've got your established zones and

those aren't supported in the neutral structure, uh, kind of changing that rule set and kind of catching people off guard. It's interesting cuz uh, so for the most part, it's it's like you don't want to change it too much because the player has gotten a bit of experience in your in your game space. So, you want him to have that advantage, right? Because I've mastered this area, I now have like a little bit of ownership onto it. It is like it's totally cool to have something that just changes the setup so just to put him a little bit on ease. Um, as long as it's explainable, like it makes sense that it happened. So, there's he understands why it happened and it's different than what he went through earlier on. So, that's totally cool to do that. Um, Uh, I had another point on that one. Um, Uh, yeah, I forgot completely forgot. But, no, like it it it's totally okay to actually flip stuff around. Um, just make sure that like your engine supports it. Like stuff that we tried to do like um, switching things dynamically just is much harder to do in open world

space because like there's, you know, geometry, nav mesh and what not to to consider. So, um, it's the you have to deal with the technical difficulties about doing that. Um, Ideally, what you want though, uh, and this is the hardest thing to do technically or memory wise, is that if there is a change in the world, you should want to save that game state so that when you come back, like there's memory of what the player did in there. And so, like one of the first ones I remember is like playing old GTA games where I had a mission and the whole building like went on fire. And so, every time I came back that area, I was driving my car and had This is the building where blew up. Uh, and so, creating that like that memory space inside of your game is also pretty good practice. Cool. Thank you. No problem. I have a uh, question about Assassin's Creed: Black Flag. How exactly did you guys do the layout for the islands and like the distance in between islands to where you keep the player interested,

Mhm. but you also give them a chance to just zone out for a bit while sailing. So, I didn't personally work on Black Flag. So, it's I'm going to guess. Okay. Um, but it's the same thing as what I was saying earlier on where like you create a like a discovery grid where you you space out your main objectives between things. And in between those objectives, you want to distribute uh, opportunities or points of interest on a fairly regular basis. When I I on Brotherhood, our general rule was like you want to create uh, opportunity points like every 50 m on a grid. And then I Black Flag is probably bigger because especially in the naval stuff, right? Like your speed of travel is much bigger. So there's probably a correlation to be done with like amount of time I need to move before encountering something else as well as the mini map. So making sure that you're teasing these opportunities on that mini map so that any given point I have like two or three things that are there that

can draw my attention. And don't go overboard cuz at some point like it just becomes like icon soup. Okay, thank you. No problem. Cool. All right. One more? So with open world, you kind of often have to reuse a lot of the same systems for people to understand what they're doing and uh which can often maybe lead to boredom which is like irrepetitiveness. Mhm. And do you have any have any thoughts on how to like balance um the player knowing what they're doing versus being bored of it like So it's a good point. So it's something I'm I'm still struggling with in general like trying to find the right formula to like expose the player to these things. Right now I'm sort of like there's there's there's a couple camps, right? There's the camp where just give everything to the player and let him go at it which is good. I mean that's it's an open world game, right? Like you want the player to have access as much as as much of the content as you can. However, the risk you have with that one is he

can get familiar with a lot of it really really quickly and you're left with no more cards in your deck. Uh so there's nothing you can surprise him with. So my current standpoint on that one is to just yes, put a lot of cards on the table when you start, but keep like at least half of your deck hidden so that as you progress throughout the game as the the player like gets more skilled, you don't have to do it like on a linear standpoint. You can do it like in branches. So like the all of the the stuff from the Skyrim games like where you have the major skill what that like depending on how far they are in there, you can then start introducing new elements or new gameplay features in there. But I tend to want to hide some stuff that I can just gently expose as the player goes on. And so you're always giving him surprises. A kind of follow-up question is uh when do you think if you have categories for missions very clearly set up like how you how you develop them? Mhm. Uh do you think that can affect the repetition of things if you have them

very clearly set up? Like these are outpost missions. These are assassination missions. But this uh This might be a point of obfuscating that in development in some way or Um I think like as long as you have enough tools to create variety on top of that, it's not a problem. Cuz like patterns are good patterns are your friend. Players like to see things that like oh I know I know how to do this. But then inside of that this inside of that content so the outpost for instance, that's where you start playing with the ingredients I was mentioning about earlier on and layering them to create surprise and and and differences. So play with the objective type, but having a couple like recurring gameplay objectives is actually a good thing. Thanks a lot. No problem. Cool. Thank you all.