Meaningful Choice in Game Level Design

Thanks I hope you guys all enjoyed the tutorial today I really liked some of the talks even though it wasn't even full attention they completely hit home actually thank you for making a decision sticking around because that's what this talk is about this photograph you've been looking at which is my title slide it's a photograph of three wheat Warbler eggs which is a bird and one cuckoo egg and I find it intensely beautiful but it's also not back to my past because back in 1998 I was studying biology at the university of essence my hometown in Germany and I realized I wasn't really into it anymore because as much as I loved biology I like playing games more and I realized that all I was doing at night which was to make doom and quake levels and hanging out on IRC talking to other people doing the same stuff I realized I could actually do that for a living and I could become a professional game developer now at this point I was already an accomplished level designer and that I've been making quite a few doom levels which you know got pretty good attention were included in some best-off list and I contributed to what we called megawatts which were these high profile community map chests that we would give out and they were complete 32 level replacement and then I started making quake levels as soon as it was released because just like everybody else I was looking forward to quake and I wanted to be ready which became a crowning amateur achievement I guess it was called beyond belief it was a nine level single-player replacement for quake and I got a lot of attention it was featured on blues news which was my crowning achievement back then it gets like 10,000 downloads in the first week which was awesome back then and to that I got my first job offer in the gaming industry virtual entertainment because some of the people I knew from IRC championed my work there so pretty much exactly 16 years to the day in March 1998 h22 I found myself at the airport with a single suitcase making the decision to leave friends family and country behind because I was moving to Texas Dallas Texas so since then I've worked as the level design on quite a few games and the last game I actually created levels for was in an official level design this was dead space too but as she changed gears a little bit afterwards I was the lead designer on Star Wars 1313 which unfortunately got canned right after last GDC because Disney doesn't like making stuff and now it's all good right I'm actually the design director on an unannounced I can't really talk about it buy a shoe totally awesome project at 2k games and initially this was kind of bothering me a little bit because I was hitting this roadblock about what to talk about it this GDC because I haven't shipped anything for three and a half years and I didn't know exactly what I could talk about but I realized that there's actually some overlap between the last game I worked on which was dead space 2 and what I'm doing right now or I have been doing in the last on last couple of jobs which is being in some sort of design leadership position because in those positions I have to be able to explain what I like about games right it's a design director I'd better be able to formalize some preferences about games into some sort of rules because I want to be able to put that out there discuss it with the design department and so we can create some sort of shared design sensibility for the entire project and I realized that the closest I've ever actually come to doing this at least publicly was in 2012 right here at level design in a day because I did this talk called player stories and designer stories which was framed around dead space 2 so in this talk as a super quick recap I explained my preference for the player story because in my mind that's the biggest way in which we use the medium specificity of games and I propose that strong player stories are found in games which are rich in agency so I talked about agency alright I explained how games actually afford agency in multiple areas systemic spatial scheduling agency and then went into exactly how systemic agency is the only one that's actually fundamental to games the other two are optional but we need to have good systemic agency to create a really good game play for good gameplay experience so saying that this kind of game is still a great game right even if it doesn't allow the player to choose multiple paths to the level or to customize the character this can still be a great game now all of this was just in service of refuting one specific map that you see on the internet a lot which is this one because you often see this described or to describe how modern first person or third person games are crap right oh my god everything was better back then and look at deulim back then you could explore and now you can do it and I was just pointing out that this is actually incredibly misleading because a game like that space 2 which might be on the right there can be just as fun and the for just as much agency is doom on the left there you just wouldn't be able to tell it from the map that's because the map doesn't actually Express everything that's going on and the systemic agency so that then manifested itself in this little comparison chart in the end I was saying you know what both doom and deadspace actually fought the same amount of systemic agency they're quite similar in the complexity of the low-level gameplay interactions that they create so they're both as much fun and then I said that this map needs to die because you know linear levels can still create a short a sense of really strong personal player stories so that was in 2012 bringing me back to today and a problem I have I like this talk I did in 2004 I think it was actually really good just hasn't helped me at all at my job it hasn't helped me add expressing a design opinion and that's because the talk as I found out wasn't really actionable because I didn't explain how systemic agencies actually created so I can't use it as a reference to talk to other people about it right and I was pretty much saying in 2012 that the core gameplay of doom and Dead Space 2 which are both combat she has some major commonalities because they both afford lots of systemic agency is just an oops if you asked me to elaborate I couldn't actually tell you how now actually know why this happened for one it's because agency the actual topic of that talk is a result it's a player facing definition of something it's a player facing different definition of players feeling empowered so any analysis of this is bound to not be useful to designers because we're describing something that's important to players but there's a bigger reason here and those were three unspoken assumptions that I realized I made that I shouldn't have made now in my mind it was always crystal clear that the way we create agency is well it's going to be a result of meaningful choice so did even go into this I didn't elaborate at all I just assumed that well duh if I tell you that you can take multiple approaches to combat space like I did using Bioshock as an example everybody would automatically know how and why those approaches are meaningfully differentiated and how its meaningful in the first place now the second assumption was that in my mind meaningful choice had been really clearly defined through words of play actually so I didn't I didn't actually explain that concept because I figured it was a shared understanding what I can do this now actually real quick and then the talk is over with of plays definition of meaningful choices that I am that meaningful choice has to be discernible and integrated in other words any choice that we present to the player in the game needs to be obvious to the player both in the setup but also in the outcome that we get of it and it needs to feed back into the game and advance the simulation in some shape or form alright and I really like this definition I've been using a definition in my own little design razor like if you work with me I usually analyze and design good gameplay by just asking what is the skill check or the meaningful choice the idea being that if we can really dig into that at the end we'll kind of know if we have something that might be fun in the end and okay great the big problem is assumption number three I was actually making the assumption that this was a definition which was widely disseminated and agreed upon and that's just simply not sure as I've been looking through all of these game design psychology biology books and so on the word meaningful is all over the place and it's always used to describe slightly different things without ever really defining it I mean I didn't even bother to put examples up here because there were too many just believe me I didn't know exactly what meaningful was supposed to mean and I think that's a real problem because now if you want to talk about meaningful choice you don't have a shared understanding as to why decisions matter and what meaningful choice actually is and when meaningful choice is and isn't same goes actually for this famous quote I mean it's clever but it's not actionable either right because interesting and meaningful are pretty much the same thing so how are you going to act on that so we don't really know what's interesting or meaningful and when we don't know if a decision is meaningful or interesting and we're not we're going to have a whole bunch of design meetings where we go let's do that because that's totally interesting to the player and that's not really good design practice to design anything around and spend I don't know how many million dollars on so that's what this talk is today I want to talk about meaningful choice in game and level design and instead of making assumptions this time I'm going to start from scratch so in this talk I want to fully understand when and why choice matters to us I want to understand why we play games as a result of that and then I want to understand what systemic agency is because once I know how a systemic agency is actually created and hopefully you guys know as well what I'm thinking about this I can then go back and I can qualify why I believe that both doom and deadspace are really made from the same cloth after like all these years apart and then finally the reason that the title of this talk specifically implicates level design is quite simple I learned everything I know - my work is a level designer that done especially making doom level so in the second part of this talk I'm going to be talking a lot about doom and about how gloom combat works why it works and how it creates systemic agency and why I believe that even after all these years doom can still be a role model that other games can aspire to because we might actually have forgotten some of that now before I do this though I'm gonna do some fundamentals I'm going to talk about games in the second part but before this I can imagine an alternate reality in which I never actually made the decision of coming to America so I'm back in Germany I'm a biologist now and I'm trying to figure out the same kind of question and why do decisions matter to us and because I'm a biologist I might actually dig into evolution and I might try to understand the biological underpinning as to why decisions are meaningful to us nope nature actually creates some very strict boundaries as to when and why organisms make decisions and understanding those boundaries is actually going to be quite illuminating but before I get to that in just a second I want to give you some perspectives as to why nature is such a hard as in the first place having blame is a luxury grains are really expensive might be lost on us because we have a McDonald's around every single corner so we can just get all the energy that we need to actually feed this really hungry brain but 20 percent of energy for organ that's that tiny that's a really big problem so in nature having a brain and using that to make decisions really is a luxury they have actually more and we also can't assume that a lot explained automatically comes with benefits because there's been some really cool artificial evolution in experiments with food flies that show how being smarter doesn't actually make you fitter to survive like they read some really smart food rice and have some really dumb food flies and then put them together and all the smart food food flies just died out because there was some cost to actually being smart and actually thinking about stuff and I guess there's something to be said for intuition / conscious thinking so there's a real physiological cost that we know about even if we don't know exactly how to having a brain and that's the cost of building it maintaining it and using it and when an animal does have a large plane and then use it uses that large brain to actually make decisions there's got to be a reason for it now there's one more major cost to having a large brain and tell me why that is but it's the cost of making decisions now of course the problem of decisions is that as soon as you make a decision you can make a mistake and nature doesn't like mistakes mistakes mean you're going to die and nature is going to try to protect animals from making decisions in the first place whenever possible and is going to try to give them some sort of innate behavior so in biology there's actually an entire field of behavioral ecology that studies exactly when and why and how animals make decisions now I'm totally butchering it for this talk but the general rule is this there are three situations or there are two situations in which animals learn and there's one where they don't learn and that's one where they do you like that transition like my eyes always glaze over so animals don't learn when a pattern is so simple that an innate strategy is possible because in this case evolution can just slowly hard-code the entire reaction via an innate behavior so we get a cheaper brain out of that animals also don't learn and they don't make decisions when a situation is so complex that a strategy is completely impossible to discern because the actual patterns underlying that situation are not discernable so we might as well act randomly right that's just as good as a strategy if your little animal living on the forest floor here and you can't really like make out all of this perspective you're not going to understand when the next meal pops up it's pretty much impossible for you to actually make order out of the chaos so you might as well start randomly because it's just as effective and it's going to be much cheaper again so this pretty much nature of button mashing right because that's going to get you to the same result now this doesn't mean just to be clear that there isn't a pattern here it just means that for this specific animal for its particular lifestyle it's not possible or worthwhile to actually start making decisions now of course there's the spot in the middle and this is where we are making decisions it's what I call the sweet spot it's an environment that it's so variable and complex that we can never actually develop an innate strategy for it but where the patterns themselves are still discernible enough to where we can actually learn the patterns and we can start making decisions and as a general rule in nature this happens when your own best strategy is not your children's best strategy but more generally you might also say that the more unpredictable yet discernible an environment of a species is the more decision making is going to occur which of course is exactly our environment right that's the environment we live in and that's what we like learning so much because we live in this world that's we do that so we can make future decisions now we life is a rock so the reason why we play games because real life is so seemingly chaotic that navigating all of the uncertainty of the real world is really emotionally taxing all right real life is not so complex that's a problem that we might as well just randomly I mean that might be a strategy I don't know if anybody's tried that out but there are enough patterns to way actually want to learn and make decisions but the problem is that it's certainly not so simple that there's one innate strategy to deal with all of life well that's one I guess which is heavy drinking but beyond that there aren't really innate strategies to actually dealing with life so the wheeled world is much closer over here somewhere right where it's a Sur Noble but not always like we can't actually always make it out and it is to this now games on the other hand are exactly in the sweet spot because they're unpredictable yet highly discernible environments which are really rich in feedback so achieving mastery over this environment is just immensely pleasurable to us because it's the better version of life so this should help us on the biological level a little bit why and when choice matters well at least the way that nature looks at it and choice matters in nature when the environment is so complex that in innate strategy is not possible but patterns are discernible and the best strategy changes across generations and we play games because the real world is often seemingly random whereas in games patterns are much easier to discern and feedback is much easier to get so realizes over there we came to assemble over here now finally the biologist imagist means that the least that choice matters to all of us because they're saying that a mind is a terrible thing to waste is actually true at least in nature because alou evolution won't even let you create that mind you can't waste it because Nature doesn't develop any sort of brains for frivolous actions animals only make decisions when there's no other way to deal with the situation or until they create McDonald's I guess now this is a little bit of a false equivalence of course we cannot simply equate human brains and human decision-making to animal brains animal decision makings but evolutionary biology is valid and I just this is like incredibly humbling to me just having the knowledge that they are various birds out in nature which digest parts of their brains seasonally when that brain isn't required because it's actually cheaper to break down the brain and build it back up next season than to keep it that's incredibly humbling right so the way I look at every decision that I put in my games is that they should be subject to that same kind of evolutionary pressure where it falls into this area where nature would actually require you to develop a brain for it no that's all great but it's 60 minutes in and I've been talking about a lot of biology and I actually put a checkmark into these two points but I'm not quite done with it because this might be a little bit esoteric so actually want to look at it from one more angle before I get to do them and that's the psychology of motivation which might be something that we're all a little bit more familiar with not until up until the 50s 59 actually psychology was bleak behavior was always reduced to a small number of primary drives hunger thirst sex the avoidance of pain and pretty much everything that we do in most mostly motivated by deficits so you are always in the red and you were just trying to break even you're always looking because we're hungry Thursday we wanted to do the other stuff you know and as you are doing all of this you didn't want to get hurt and that to me is pretty bleak and luckily when actually I was a problem of this Misti we could not actually explain play there were a whole bunch of examples for humans and animals were engaging in activities that didn't support any of those primary drives and it's kind of hard to justify that drive theory is the primal way in which we're actually making decisions when rats are going through electrified wire just because they want to explore obviously the avoidance of pain is not a strong enough Drive for them to just go off and like look at some other part of the laboratory so this is Robert white in 1959 came along and he put a positive spin which I liked much better on what motivates us he took intrinsic motivation seriously and he championed it as an actual primary motivator he proposed this concept of competence our motivation for activities in which the soul we worked for sponte the spontaneous feeling of interest and enjoyment that occur when one engages in the indi activities now these days we unite competence as well as autonomy and relatedness and then the umbrella of self-determination theory which is one of the most widely studied and applied theories of motivations including games there's actually a book which was co-written by one of the founders of self-determination theory called glue to games and it applies all of the findings of SDT to computer games and analyzes exactly why games are incredibly good or fulfilling our basic psychological needs and why they make us good to play now the needs again are competence autonomy and relatedness and I don't want to go too deep into it and try to really get distracted by all of the details of the psychology if you do want to do that you can read this book and you should read it on top of glue two games because there's no reason why every game design and game developer shouldn't have done that already but so here just a highlights the important things you need to know for the rest of this talk SCD proposes that all three of these categories competence autonomy and relatedness are universal needs that they are not motivated by deficits that they're in nature all human beings and that they are essential to our health and well-being and if one or ideally more than one of these needs is fulfilled by an activity we feel we feel intrinsically motivated right we feel good so games are really good at making us feel competent autonomous and related we save the world we explore the world or when ourselves as we're doing that and people keep showering us with recognition or we get to kill them in multiplayer I mean that's the way to feel related to somebody as well I guess and I think that SDT can also explain when and why decisions are meaningful to us because I submit that definitions in games decisions in games are going to be meaningful to us when the outcome of the decision satisfies one of our needs of competence autonomy relatedness this is something that you do not see explicitly explored and good to games because ultimately it's a psychology book but it has to be this way because games make us feel good by satisfying psychological need the medium specificity of games interactivity and the only way to engage with that interactivity is by making choices so making decisions must actually lead to some sort of emotional well-being but only and that's the one big caveat if our like psychological needs of car as I call it competence autonomy relatedness actually fulfilled right so this is the other way in which I'm looking into why decisions are meaningful to us if one of these base needs is going to be fulfilled in some shape or form then the decisions that lead to that they're going to be interesting or meaningful to us okay so let's get to doom and let's get to an example of how all this information hopefully all starts forming one big picture and starts working out in practice and how through all of this we can figure out what this mythical systemic agency actually is and why Doom has so much of it actually wanna do one quick poll who here by show of hands is over 28 years old okay that's still quite a few it's absolutely amazing to me that doom is 20 years old because you know what that means now you can tell me you did it but there's going to be a whole generation of level designers and game designers we're going to be entering the industry who went even a life when doom came out and that is pretty crazy but so it actually is 20 years old so you know what let's just gain some perspective and remember or imagine if you have to what it was like to play Doom back then for the first time and you're entering this 3d environment which is completely overpowering your senses right I mean yes Wolfenstein had been our Ultima Underworld was out there but the way that doom really controlled and how you were moving around or something very new and because it was very new you had no idea how to move around you weren't really good at it I mean how did you even control this game was it keyboard mouse joystick I went to all of those notes oh no you're clumsy moving around because you don't know exactly how to control the game you don't know what's ahead so he scared shitless of all the monsters coming at you you're probably just standing there and shooting at them frantically because you can't even move around competently but then you play the game for a little bit and you realize you can actually strafe and what's better you can actually strafe to avoid all of those fireballs which are shooting at you after a while you might even a will be able to good enough or be good enough to use that strafing to get closer to enemies instead of running away from them so you should actually start feeling pretty good about playing this game well then you hit a one m8 you've meet your first two parents of hell and you're mortified again you die a couple of times but then you know what you realize you can circle strafing on them as well and they're not really all that bad as I figured out because they take exactly five four blasts with the double-barreled shotgun and they're done now of course to be having a double barreled shotgun you had to be playing doom 2 because doom 1 didn't have that weapon but that's actually the game that I made all of my levels for and it just took me that long because I didn't have a modem before that so I couldn't actually download an editor but so when I started playing doom then or when I started making levels for doom I was already an expert doom player and I could actually apply all of that knowledge as to how doom was fun on that level to making my levels and it was one hell of a formative experience I mean obviously I'm doing a talk in it now it taught me a lot about gaming level design the doom is also a great example of how good action games satisfy competence needs and how at the lowest level actually make us feel good so I want to look at that first competence is defined as an intrinsic need the intrinsic need to effectively deal with the environment actually master it to not be bumbling around and falling over on feet and so on due to games calls it the experience of overcoming a challenge or extending our abilities in a meaningful way to be truly satisfying however our successes need to occur in the context of a real challenge and that's what Doom is right it provides an environment that we can master in the context of a real challenge there's no doubt that the fine-motor skills required to aim the timing skills that you're going to have have to have to line up enemies and shoot them the judgment that's going to be needed for missile weapons because those are moving enemies and just the general rhythm and cadence it's going to come out of how quickly your weapons and the enemy weapons are shooting now these are all elements which are going to combine into a very compelling challenge so the act of playing doom at this low level can be committed to procedural memory at some point what we call muscle memory right at some point I can actually play Doom in my sleep it's not quite an innate behavior but it's the closest we can get to actually having something imprinted in our brain so everything about Doom kind of feeds into an established narrative of why we like action games so much we inherently motivated to achieve competence of an environment or an object I can't still can't do this and games are these really feedback a feedback rich environment that tests all survival skills and I keep us optimally challenged as we're achieving mastery over all of the dangers which are literally like fretting our very existence now the question here is that the way we talk about dunes we're talking about survival skills does that actually have to do anything with choice like I said earlier that decisions are bound to be meaningful when they fulfill some sort of base psychological need like competence but how does this actually apply to skill check survival skills because this is supposed to be instinctual right supposed to be imprinted in our muscle memory so where the lowest-level is the choice here maybe my entire thesis just falls apart and I shouldn't have done this talk but I mean that's part of the narrative Doom is a really good action game action games are fun and compelling and meaningful because well tens and oh my god look at all the skill checks but in fact if you're looking at my designer razor I have explicitly put skill check as a component in there so I can actually test for that like the twitch skills which are going to come out of something now one thing we have to remember here where it doesn't fall apart is that the skill check is not something a player actually does it's something that a player reacts to and how we react to the skill check is by making decisions all right according to glue two games do two games a game has to have three properties to fulfill these competent needs it has to have clarity about the goal at hand challenge that is not overwhelming and clear feedback and our actions that make us feel like we learned something and it's exactly this last part right here that is so important because it actually does explain how games are meaningful at the moment-to-moment level because if we do react to skill checks by making decisions and adjusting our aim and so on we need to be fully in control and that requires very good science and feedback as well as very good controls in the first place and so what I really mean when I ask what is the skill check or the meaningful choice is probably this how good are the three C's and the science and feedback that are implicated whenever the game expects me to react to a skill check by making a split-second decision right because I can only make split-second decisions really well if the three C's camera character control are awesome and when the sense and feedback I'm getting to everything that I'm doing is very clear and unambiguous because then I can get into this flow state where the controls just meld away and I'm one with the game and that's where my competent needs competence needs actually fulfilled because the amount of immediate feedback from the game like to each of my low-level actions actually allows me to master all of the skill checks that the game is throwing at me so science and feedback at least in the skill check sir now you are actually the lowest level in which a game creates systemic agency and agency in general and that might be a little bit weird because the way we usually talk about see is in terms of big decisions and big choices in fact we often abuse it now to just say something like player freedom which is wrong agency just means at the simplest level that it is the satisfying power to take meaningful action and see the results of our decisions and choices and that's exactly a big part of the reason why we feel so much in control of the combat encounter and we get into this flow state in a game like this because of the low-level systemic agency that's afforded via the science and feedback and the three C's and that then in return create satisfaction of our competence needs right that's what we're feeling really good about having maths at this environment so you might actually say that doom at the lowest level is is the use of interesting decisions although we might really have to say it's a series of low-level control related decisions but everything I do in this game is discernible and integrated and the decisions that I make such as how to move around where to move when to shoot actually matter so that's part of this mastering skill check we are good pre-season via good science and feedback and they do need to be good because if they're not good we can't control the game and then the competence needs aren't actually fulfilled but this creates agency and this low-level agency that I'm talking about is certainly part of systemic agency and making sure that you know the game plays and feels good now of course the question is is that all because if the fulfillment of all the competence needs is the only good thing that an action game needs there's a lot of games out there that should be awesome right go to games actually as a section in which analyzes doom specifically through the lens of competence it's kind of that one of the poster child's and it makes this argument it says that it is the gradual progression of challenges as the player marches into hell that provides the strongest motivational pull of doom and virtually all successful at PS games that progression is key with each hellish opponent you fail there's a small victory one on your road to the larger goal from a competence standpoint these small moment-to-moment victories are more like a trail of chocolates each providing a small bite of satisfaction as the player succeeds on progresses towards a larger victory it's quite poetic I don't want to put words in their mouth to me this reads a little bit like what they're saying is that you need to be able to point a gun at somebody and you need to be able to overcome a progression of challenges of skill checks and if you do that you're going to have a really successful first-person shooter game don't get me wrong actually do agree with us in one way I do believe that dooms rational level is iron like the game system story of which I didn't have a picture except for 1313 but I believe this is important all right like um overcoming the challenges thrown at me in Doom and how finally that's tuned is a big part of why that game is good but having played as much to me as I have in my lifetime and just playing the same level over and over again I think if you just have this outlook we seriously missed a mark because we're trying to identify competence needs as the main reason as to why blooms gameplay was so successful and how many modern games do you know that have good aiming good movement good science and feedback and they probably have an enemy progression as well but things get harder and it's not really fun to play because there's some sort of spark missing I mean how I only have to stay within the same franchise and go one game down to know that a competent execution of a progression of challenges is just not enough so there's a danger to thinking about action games in terms of confidence only and that's a trap that many games unfortunately fall into would have left to play an extra clip here but this is why playing doom 1 and analyzing one and two is really a big deal to me that's why I'm doing it because to me it's the reminder that doom doesn't just frame its core component it's systemic agency in terms of competence there's something else to doom something else that needs to be there to really make good like good combat in these kind of shooters something that satisfies another psychological need and that's autonomy and pretty much what this comes down to is that Doom works because it has a focal unit differentiation now I love the term orthogonal unit differentiation because in pretentiousness and awkwardness it's only second to ludonarrative dissonance so i spare myself some trouble here I just call it Oh UD or simply differentiation but we should talk about it and it's something that javi Smith introduced at GDC in 2003 so this was like 11 years ago Oh UD gives us a way to talk about how game systems are meaningfully differentiated in their abilities and behaviors and how that distinction leads to some interesting and complex gameplay it leads to what I called in 2012 an expressive possibility space but once again this was a little bit misleading so I'm going to take it from the top and actually start by establishing a better presentation a representation of Bloom's whispers possibilities based in the first place here's something that you figure out after playing doom for a long time the enemies in the weapon in this game are actually really finely tuned to interact on multiple axes but first looks like this visceral barely controllable maybe even simplistic fight for survival is actually this really complex interplay of many gameplay elements that allow for ever new tactics and strategies and this all happens via oud because if you're looking at the enemy classes in Doom I can map every single enemy type across across some orthogonal e differentiated categories there's an axis where I can map all of the hitscan versus projectile dice all right those are the two classes of enemies that exist into them and then I can also do this by the amount of damage they do and then there's a second axis which are charging enemies versus enemies which stay back which is just a function of the overall speed but also the behavior that they're subject to then if I put all of this together I can create a chart that looks like this and this is actually pretty accurate representation of Doom's possibility space at least as it pertains to these specific enemies and what's important to me and as you notice the enemy distribution actually already covers a large part of the entire spectrum I left out some enemies there so I'm sure I could pretty much fill it in all the way but what this does is allow the player to have a very specific opinion on each enemy type because they know what they stand for and that allows them to prioritize enemies in each encounter so if I'm looking at this example like which enemy would you want to take out first with your shotgun I'm assuming it said black shot gonna do it back there because he has a hit scan weapon and he can deal immediate damage and I know about this where's all those fireballs that are phone at me I can dodge those those guys are called imps by the way because he didn't play doom um so the reason why doing works so well is that each enemy weapon indistinct consistent and discernible behavior that can be learned and because there are only a few trades to all of these enemies which just fall across two axes I can actually learn this rather quickly so this is a this is information that players are going to internalize and that's really important to them so I want to look at how the play actually processes and stores this kind of information and they do this via a psychological construct called a schema which is an organized pattern of thought or behavior that organizes categories of information in the relationship among them and then information which is thrown at me is either assimilated it's pretty much integrated into the schema into an existing one or can be accommodated which means that it doesn't fit I actually have to create a new schema for this information because I realized that a striped horse is actually a zebra and not a striped horse we can actually be a little bit finer with what exactly a schema is and we can do this by what information they store we call schemas a frame when they contain properties and behaviors of something and we call them a script when they when they encode some sort of processes how we approach a situation or how we deal with an object so in dooms case the attributes of the enemy would be its frame and while the player tactics of our actually approach this enemy would be its script and players can easily create frames and scripts for each enemy type in Doom they know exactly what the enemy represents how best to deal with them and everything is very clear-cut because it's all differentiated so so far so good um all of the individual units have clear and concise properties and behaviors which creates like some very clearly discernible patterns to each one of these guys and just each pattern encountered on its own is going to be a no-brainer what's interesting is what happens once I put them into conflict because then quite literally all hell breaks loose looking back at this picture right here do we really want to take out the shotgun or first is that a script that I can reliably follow for all of doom deal with the hitscan guys first because that black dude in the back certainly can shoot us immediately while I dodged all the fireballs by those imps except that that imp over there has actually gotten pretty close to me all right I might not be able to strafe out of the way anymore so this is actually a judgment call I don't actually know exactly how I'm going to prioritize such the situation now call this funnily enough prioritization choice because I get to have an opinion on how to I approaches encounter and how I prioritize it now prioritization choice is created from the complex interplay of systems that are easily understood individually but that combine into situations that don't have a consistent or obviously superior tactic and that creates a really interesting game in Doom in which the overall combat experience is clearly defined like there is a quality to playing doom there's an overall strategy to playing doom as well and I can learn that and I still remember that have it internalized it involves prioritizing enemies lots and lots of strafing around lining up enemies in very specific lines picking other white weapon for the right job and so on but what I can't ever do for doom is create a script that works for all combat encounters in doom equally and that's because the combination of multiple orthogonal differentiated enemies create situations which in which an obviously superior tactic pretty much never exists and this is really interesting to our brains right this is actually leads to interesting decisions because what we've created now is something quite familiar it's this it's an environment in which patterns are individually discernible but they combine into non-trivial situations and these non-trivial situations can be learned because they change the cross generations well at least if you imagine that each combat encounter is a new generation so an innate strategy is never possible and unlike this flow experience like in the low-level move from the mechanics that I get to have we can go on autopilot when playing Doom we always have to stay engaged and we have to keep learning and learning is actually fun a more visual way of communicating what I'm talking about here might be through these patterns because each one is easily understand understandable and discernible but once I start arranging these patterns into a specific combat configuration maybe taking distance the player into account I get this you know it the patterns are still clearly discernible but as they overlay they become harder to make out they do offer specific points of attack this one here probably where the overlap happens as one here and looks like this really good one over here but of course combat itself is fluid it's dynamic I keep moving around switching weapons and so on so once the player start interacting by doing this something like this happens the patterns rearrange themselves and the player actually has agency over that as well the player kind of turns into a DJ like the player can actually scrap the patterns and find specific new ways of attack if I'd had flash for this I could have done a really cool animation on it right but it's just a metaphor anyway but if you play doom or you remember this this probably rings too because if you remember all of the gameplay systems there were things like enemy in fighting so I actively tried to trigger that by lining up enemies very specifically and I was scrubbing the patterns to align themselves just the right way there's weapon differentiation so I keep switching weapons depending on enemy types and on distance and that's something that was called from the paint chance where select monsters can be kept from ever actually shooting at you if you keep needling them with bullets because the calcio demon I know exactly had a 90% paint chance so the toy the chances of him getting a shot off were really really small so what actually used it to my advantage I would position myself so that the calcio demon was including all the hit scanners on the opposite end of the map so now he was this living shield as he was getting rid of willed with bullets from both sides so that's kind of my analysis what do most all about if you played the game all of this might ring true depending on how much you played the game or when he last played it some of this might have been some small aha moments but doom did all of this as I said because of what I was it afforded what I call this prioritization choice which means that prioritization is possible there are multiple ways to prioritize so this choice between how I do it and choices are roughly equally weighed so that the player has options and is able to actually have an opinion on each combat encounter and what all this means the result is that there is no clearly superior combat script which exists for the entire game which keeps it interesting now it's also something that it's not right priority prioritization choice does not mean that it's a free-for-all of approaches we're not trying to create tons and tons of trivial choices like how these choices are differentiated still needs to be meaningful as well and I'll talk a little bit later about how we offer it but I think it kind of happens automatically and actually one more thing here is what it's not well I'll get to that in just a bit let me just figure out why prioritization choice I always pick this one word where I just completely let every talk but so let's figure out why prioritization choice is meaningful and the short answer is it's meaningful to us because it makes the player an active participant in how Kombat shakes out it satisfies autonomy needs an autonomy is actually a little bit trickier than talking about competence because it's not just what you might think which is freedom it's not about just allowing to do whatever let those are ads autonomy is actually about choice and opportunity and reflect our innate desire to take actions out of personal volition and not because we're controlled by circumstances or by others it's an incredibly powerful psychological need are we more so than competence because what we don't like being controlled and which actually being threw back to this last point of what prioritization choice is not it is not a way to force enemies on the player because autonomy is a fickle thing just like its fulfillment can really improve enjoyment of an activity if you stomp on it it can severely damage motivation alright there's actually tons of research on that which you might have heard about in the general domain of extrinsic motivation and achievement and so on honestly it's game players we can probably easily just identify with the sentiment because we play games because we don't like being controlled by circumstance or by other and so we need to be a little bit careful but we don't like being controlled in a combat encounter I might argue because if authoritative enemy scripting creates something we're wave after wave of enemies it works in exactly the same way checkpoint we start after checkpoint reset as I died that sucks I mean that's that's a pattern which you know is not really something that interesting to me gotta take this with a grain of salt of course it's not an absolute rule I'm not saying that we shouldn't use attention grabbers like charging enemies to spice up an encounter right I'm just saying that we cannot spoon feed enemies be a true authoritative scripting so prioritization choice is meaningful because it satisfies autonomy needs and because it creates an environment that looks exactly like the sweet spot right here and doom has really strong systemic agency because I can actually put those two things together I have my low-level ability to master skill checks and I add an autonomy level on top of it via oh you D alright we add autonomy to what at first glance is a linear game even though the way you go through the game looks exactly the same on a 2d map the way I can play it is going to be very different as I'm playing through it again and again and again so finally the reason why all of this is meaningful to us is because good systemic agency once I put all of this together it creates meaning because we get to learn like hits we are a constructivist approach this is pretty much what Piaget was talking about right it's just this idea that the amount of information that I get to learn in combat in the game that has this really big possibility space is so diverse that we're constantly assimilating or accommodating new information we constantly update our mental models things don't get stale so would be remiss if once again I didn't talk about how we actually offer for all of this like how do we ensure that this level of systemic agency I'm talking about he has actually created now I could probably buckle down and do math furiously I could try to take all of the meaningful combinations or figure out of the number of all the meaningful combinations by telling up all of the different core mechanics of the game and of the system and then I could figure out all the combinations permutations binomial probabilities of the game and just sounds like design masturbation doesn't it I've never actually done this in my professional career beyond like sets with small cardinality I don't yeah if you really want to deep dive into this that's a rabbit hole you can go down you can look up the topic of computational complexity and you can look at how our minds actually have a preference for a game switch our pspace-complete which is like a certain computational problem with certain computers which can solve it Rothko star talked about this in his last GDC next lecture so if you want to look at it that's great but to me I think I'm going to be much much simpler in my advice I'm going to tell you that you simply have to follow the rules that we've already laid out was all there the entire time you have to ensure that the enemies are differentiated across a scale that's clearly defined on each end and those should be opposed in some shape or form ideally you have to make sure that each system has a clearly understood frame and script so that you create clearly discernible patterns out of all of this then you need to make sure that while the game controls really well and your science and feedback are clear and unambiguous there's no question when this guy is dying right just like that doesn't matter if it's a cover shooter where you could just go back behind cover he's dead and then the way I usually test this is something the super scientific method of what I call the check point test it's this idea that I can simply play the encounter multiple times imagining that I had died multiple times and I see if the patterns rearrange themselves meaningfully different to where actually feel like let me try that again because there's got to be a better way or a different way this time to figure this out and if you do all of this then he create this mystical cream that I was talking about in 2012 you get to fill your game with cream which we didn't do in unreal 2 at that point as he might remember so there's an entire list here which I'm going to kind of skip over because I feel like I explained most of them the highlights pretty much are this you need to ensure that your three C's are top-notch because we can't feel competent if we can't control the game need to make sure you have very clear science and feedback because they actually create agency at the lowest level and that's vital to a feeling of competence and you need to create clearly understood systems and weapons are part of this as well right like there's a whole nother talk where we can look at weapon design and how that is meaningfully differentiated and why there's a reason we only have like four or five weapon types across all games and everything else once again to me is design masturbation because telling me yeah you can create your custom weapon just create the system soup in the middle somewhere there that's a different talk I mean I talked about it in terms of enemies and that's what you need to do make sure they're orthogonal II differentiated because that creates the potential for some sort of prioritization choice and I hope after all of this I don't really have to go into much detail as to why Dead Space 2 is made from exactly the same cloth as doom well unless you didn't play the games but like Dead Space 2 has a cast of differentiated characters with very distinct frames and scripts good silhouettes as well so I can immediately tell who is who and that's actually their differentiated across enemies that rush and enemies that are shooting at me from behind and then it actually does something which pretty much no other game does it embeds meaningful choice right inside each enemy type as well because you get to dismember all of these necro moles and you can do that in multiple places and depending on where you do it actually get different reactions so as meaningful choice there plus you get a wide variety of differentiated weapons that also have very specific profiles and behavior so that's phase three actually creates prioritization choice on multiple levels for your enemy dismemberment enemy prioritization and we have weapon choice as well and then it adds some systems like stasis on top of it where the player then becomes the DJ and he gets to scrub around and not exactly which pad and he wants to create no he wants to freeze so maybe I told you something new maybe I told you yesterday's news because if you like all games do this to some degree I don't think we doing it enough and we might not even know exactly what we get when we are doing it and that's why I called unreal 2 was such a big disappointment in 2012 and the heroine ik part actually is that I was constantly talking about doom we made we made Unreal 2 in 2000 1999 2000 so I was constantly talking about doom and what did how awesome it was and then it didn't actually have the tools to convert any of that knowledge into gameplay but I feel that even when you do have differentiated enemy types there are quite a few games that do that you have the shotgun I have the sniper and so on there's still quite a few games that then fail a checkpoint test because they're way too authoritative and how dish out those enemies maybe they don't have the systemic depth to support systemic behavior systemic combat or maybe that's just you know what the designers are doing but ultimately my feeling work for autonomy is then stomped upon and if I don't die in the game it's going to be an awesome experience but that's just because I don't know what's missing what are some games that do it right well halo is the big one right and I don't even know how much I have to say about it there are differentiated enemies across strength and behavior with very clear silhouettes their enemy weapons their hitscan weapons there's this interplay between how I take down shields and enemy energy and so on and they actually add something which I don't think it's a lot of credit for being a design innovation instead of just a controller limitation which was weapon shot weapon shown like back when Halo came out it was the first mainstream game where you could only have two weapons at the same time and that wasn't because they were maybe was also because they couldn't have a big inventory but really it was constantly forcing you to make meaningful choices about what kind of weapon you were taking into the next part of combat and if you fail the checkpoint well how you would approach the combat the next time we played it as absolutely no accident that the reason I know about this book the one you should read is via Jamie Gries now who was the lead designer on Halo 4 the first I think three games I think Batman is really good it creates already small gameplay molecules with differentiated enemy types alright so as I'm interacting with one of these bubbles of combat I get into the slow state of combat and then I'm actually trying to prioritize so just avoid guys as well at low level and you know when somebody's picking up a weapon I'm trying to go for him so on so that's all happening for exactly what we're talking about here and then I guess that was 1313 did a pretty well and I can't put any slides up here but I feel like the oud in our game which was a cover shooter was actually quite good where with just four enemy archetypes we were actually able to get something that didn't that didn't fail a checkpoint test because we had enemies with clearly discernible behaviors where I could have an opinion on what would happen I knew this guy would charge me I knew this guy would be throwing grenades I knew this guy would not stop for anything but I had a certain weapon that I could use to slow him down and so on so that for me is just the way I've been trying to apply all of these lessons to the way I'm making games mmm yeah I'm going to let you steer this picture for a little bit longer it's actually I feel the perfect symbol for decision that's meaningful in nature I just had to cut the about 20 slides that explain it but if you want to know how the weed war player has to make a really meaningful choice I can predict when a cuckoo egg is in its nest or not might put it in the slide downloads but there used to be a time when everybody in the industry had played doom that was pretty much the baseline all right so it wasn't any question about like what it does and what it is and what it isn't and that won't be the case anymore very soon and that means it's up to all of us to actually continue that tradition we have to continue to make games that take its place right because we have to create a new generation of game and level is an asset understand simply by playing the game the virtues of good gameplay and creating this kind of systemic agency now this may have been a systems design to Arkham probably was it was but the fact that I learned all of this rule level design is incredibly important to me because me being a level designer on a really good game with low level systems made me a better game designer working as a level Azana in Doom was incredibly fun it actually still required a lot of thought it's not like you just randomly distributed enemies it's just that you had so many choices and how you played that possibility possibility space and just kind of distributed across the map and in how you then form the entire environment around that being a level design on dead space was actually very similar because in dead space the systemic agency is so high that I can create a compelling encounter by just putting three slashes in a room which is the base enemy because they're all going to advance at me and they're going to kill me there's no question about that so I'm going to stasis the first guy I'm going to try to take off the legs of the second guy so that he doesn't get to me I'm going to kill the third guy just as he gets to me at that point the first guy who is in stasis anymore is starting to charge me now oh oops I missed and actually took off somebody's head and all those guys are charging me and it's just that as a base encounter is something that's inherently interesting and just imagine what you get to do on top of that by combining all of these enemy types and actually scripting combat or on that so I didn't actually do any of this I'm not taking credit for it right I was just a level designer applying dead space as a possibility space I got to be this is what I said in 2012 I got to be a guide to the possibility space I got to make Guyver new gameplay situations from the game because the actual underlying systems were very simple and clearly delineated I don't know if you guys saw this Kotaku article which just came out like a few days ago it just made me chuckle he would gamer surprised how to HD even to find that maybe there's a reason why enemies and games are dumped in predictable well no then dump and predictable I hope you understand this because games are the antidote for a life that is too chaotic life is over there and games clearly live here doesn't mean we can't advance in a whole bunch of interesting and meaningful ways there but ultimately there's a reason why all of these things are very easily categorized in our kind of stereotypes life is however orthogonal II differentiated it might be incredibly chaotic but there is a clear choice that I get to make right I got to make the choice whether I wanted to be a biologist or game designer and I didn't have the benefits of safe games or checkpoints so in life choices matter just because they matter I know I made the correct one right I don't regret the choice of becoming a game designer one bit I just realized that just like in life game design is an uncomfortably chaotic it's really hard to find patterns of good game design in there it's very intuitive now we have to remember just in game design just like in life things only appear to be random actually are patterns in there which is why we're searching for them which is why all of you are coming to GDC right we're searching for patterns to bring some order to the chaos which is game development now on this talk I tried to figure out why action shooters like doom at the low level at the core level are inherently meaningful to us and that's because there's some puzzle box with just the right amount of danger when that's via systemic agency which is one specific pattern which I want to present to you I hope it actually came through clear I hope you found it useful and I hope that as you go into the main conference you find a whole bunch of other patterns that you can apply so thank you for coming today from all of us and that's me

Channel: GDC

Views: 73,505

Rating: 4.8301888 out of 5

Keywords: gdc, talk, panel, game, games, gaming, development, hd, design

Id: BEF4GVNzkUw

Length: 57min 26sec (3446 seconds)

Published: Tue Dec 15 2015