Design in Detail

[Music] the joys of a Friday afternoon tuck my voice is shot already and you all should be on your way to catch an airplane and we're all a little bit hungover so let's get started and then they can tear this place down so my name is Jamie Chris Berg I've been a game designer for about 15 years I originally started at Bungie which was acquired by Microsoft left Bungie went to sucker-punch that was almost immediately acquired by Sony so I'm a good luck charm if you're looking to be a bot by a large console maker I've worked on a bunch of games all of which involve colons and my my talks are very fast it's sort of like drinking from a firehose so if I'm making a point that isn't relevant to

your game just wait 10 seconds and I'll get onto the next one hopefully a couple of them will will matter to you so two years ago I did a talk that was inspired by this painting it's a Sunday afternoon on the island of the ground shot and it's an example of pointillism which means it has tons of tiny little dots that make up a enormous like 30-foot tall painting and the idea was that I would examine one tiny little detail about a Halo games and spend an hour talking about it and get really really into all the way into that one decision so I picked a sniper rifle reload and then over the course of the of developing the talk it really became about balance and how a game designer can balance the game so that once it's out in the wild they don't have to touch it anymore and it can just continue to have a life of its own and the community continue can continue to play it so then

the next year they asked me to do another talk in this series so I looked for another painting to inspire me and I picked a haystacks by Monet and when he did was he took the same exact stacks and he painted them at different times of year in different times a day and so that gave me the idea of like let's take one detail and look at it in multiple games in the Halo franchise and see how it changed so I looked at the modes of velocity and the plasma rifle bolt and tracked it across the the first three games of the Halo franchise but really that talk was about tuning and how you can change small aspects of the game to create a certain experience and then if you change that to be something else and changes the experience that it generates so in my continuing effort to wage war on the formatting of the GDC website by having increasingly longer

titles when they asked me to talk this year I said that's great but I don't want to talk about Halo you guys are sick of hearing about Halo I'm sick of talking about Halo so I had to come up with something else to talk about this year and again I turn to art for inspiration this is actually my favorite painting it's the Annunciation by El Greco it's actually really really tall so if you're staying in front of it it looks more like this and it's about the angel Gabriel coming to Mary and telling her that she's going to conceive a son and that his name's gonna be Jesus and you guys maybe know that story and what's really interesting about this painting is the is the faces and the detail and the colors but what's more interesting about this painting is it's not the first painting called the Annunciation there's actually dozens and dozens and dozens of paintings from hundreds and hundreds and hundreds of years all of them painting this one exact moment and in capturing in their

own style so that gave me the idea to look at different games that address the same problem maybe had the same details and compare them so this year we're not talking about my games we're talking about your games I'm sort of infamous for for a blog post that I wrote where I gave uncharted to a 7.5 and then I gave jumper a game that none of you have played also a 7.5 and really the point is that uu is a game design you should be playing really good games and looking for the flaws and looking for the problems and you should also be playing not-so-good games and trying to draw out the promise and train yourself to find the gem in the rough so that means if I'm criticizing a game that means I think it's a good game so I hope I'm not burning too many bridges in this talk by criticizing a bunch of games from a bunch of very talented developers and

designers keep in mind that I'm speculating I didn't work on this these games that actually made it a lot harder to kind of draw out some of the details so I'm doing a lot of guessing and if I guess wrong about your game don't be insulted if I guess right by your game just don't tell anybody and they won't know so welcome meant that to the actual talk which was so long they couldn't even fit it on the placard outside but before we get started let me talk a little bit about details and why I think there is so important if you change just a 5% of your DNA you go from being a human to being a chimpanzee if you're leaving England for the new world you get a couple degrees off course you end up in Massachusetts and not where you want it to be in Virginia so small details especially early on and have a an impact down the road we live we work sorry in a

fairly mature industry now there's lots and lots of professionals out there I'm on great teams they're working in established genre and what that means is that a lot of times a really good game and a not-so-good game have a lot in common there's very little difference between halo and haze for instance but there is not quite as similar an outcome when it comes to Metacritic scores or profit and details are the things that make the difference between success and not so much success it's important to get the details right another example of how details are powerful is Angry Birds so Rovio made like 50 games before they got to Angry Birds and that allowed them to really nail the details of working on the iOS platform it's it's it's got great controls for a touchscreen the time investment is quick enough that you can finish a level in the bathroom

it's the reward system is is super gratifying and actually even arrived on the scene just with this perfect timing and they're definitely reaping the rewards on the other hand a game that just recently came out I'm charted 3 but again Naughty Dog is like the Pierre the peak of the industry right but they they missed a couple little smile minor details like the the camp controls I feel like the animation cycles got a little bit longer so you feel like you're a little bit less in control sometimes the the jumps can actually take you farther than charted 3 so it's a little bit harder to wind your way through the environment and so they really they really suffered the cup and no no I'm sorry they reap the rewards to you never mind they got every other detail right there 5 the difficulty comes in the fact that details are so subtle so who here thinks limbo had like fantastic art direction ok I want to I'm

an arguing wrong what they had was great design direction the fact that everything's in profile is what makes the game work at all I mean is that a bear trap or is that some grass if you could see it and it was just rendered it wouldn't be much of a puzzle is that a spider coming over to get me or is that just some branches in the background without the uncertainty caused by this detail of the rendering the game wouldn't even be possible and the best part about details is they tend to be portable so Darksiders is well known for liberally porting details maybe to their engine so like the hookshot or boomerang or the portal gun but because they did such a great job of capturing only the important details and changing everything else instead of feeling like copycat design it really feels like a Greatest Hits game where you get to go through all the best game design details of the last decade on the

other hand if you don't know which details are important you end up copying too many details and your game is going to end up feeling like a Xerox machine that got left on you'll get kind of pale imitations and if you really don't have any idea what you're doing you have to copy every single detail and you get caught and that's a bridge that I'm happy to burn so the the most important thing about details when it comes to being a designer is that they're manageable there are thousands and thousands of decisions that go into making a game and there's no way that a game designer can keep a handle on all of them so if you were a ratatouille this the the little chef the rat couldn't couldn't cook in an oversized kitchen right but he could grab his his

friend's hair and use that to control him and ended up making great dishes so in this metaphor designers are rats usually the players are there rats but in this committee for the designer is the rat and the game is the the larger context and the the hair is the details I suppose so that leads me to the real kind of theme of this talk which is that before you can balance it before you can do the details that would make your game successful you have to identify them and so what I've done is kind of gone through my design process and tried to hook up the details that I found in playing all these games to different aspects to try to give us a little bit of structure Eric Zimmerman did a talk yesterday where he said we shouldn't in from instrumentalized our game design and so I wouldn't make this clear this is my design process I'm not saying this has to be everybody's process the first step is to get a concept and concept are by definition undetailed

right they're their notions they're this like here's kind of what I want the player to experience but it's always vague at the beginning so I'm not going to go into any details about concept details so the first area that we're gonna explore is constraints if you're familiar with like the MDA concept of design they would call this mechanics but for me instead of thinking that as something that you're building and then when it's broken you have to fix it I like to think of it as constraining the player and then when something's broken I just constrain them a little bit further so one way that you can constrain the players by removing options God war is a very successful game that got there by removing all camera control which is which was sort of new for the for the third-person action game which is great because the game is all about stabbing guys and cutting off their heads at doing horrible things to them and so not

having to worry about the camera actually freed you up to do what was good about the game on the other hand ocarina of time yeah we're gonna go there automated the jumping process and for a game that's about exploration and jumping is like the one sort of vector of exploration that you have not being able to jump meant you couldn't explore as much as you want and could be frustrating another way to think about constraints is as a cornerstone when you start a building you you lay the first stone and that kind of determines the rest of the building whether it's gonna be square so in Halo our cornerstone at least for combat was the engagement distance so we decided how far away we wanted enemies to be in that meant we had to have a pretty slow moving speed because we didn't want you to be able to run and get there right away we had to have a pretty narrow field of view because we wanted enemies to be large that engagement distance but I determined how accurate our weapons could be the determined how many enemies

we needed to have you fight at one time it even determined the AI firing primers because once the AI says frightening you you stop you start firing back that is where the engagement distance comes from so in halo the engagement distance really determined the environment scale but if you look at game like red dead they were clearly trying to create a vast huge environment right so their movements being had to be relatively fast because you had to cross very big spaces and their field of view had to be very broad because they wanted you be able to see as much of the environment as possible and then with their weapons had to be really accurate and that they're their enemies were spread over a very large space and it also meant that the a I had to fire pretty repeatedly because that was one of the only ways that you could find them by the giant flash from their gun and so they achieved the environment scale but because they determined that detail first then their engagement range you know in my opinion suffered but this

isn't intended to be a criticism of their priority just explain how no your priority and then and then know that that's going to affect the final game one way to think about this is solving it for gameplay so computer role-playing games have this advantage where they're using computers they can do really really big numbers and so they tend to equate having a higher level at doing more damage and the problem is that this leads to this exponential curve where if you're doubling the damage every time you go up a level at some point you're going to just a few levels after you go through content going back through it's gonna be completely trivial on the other hand Dungeons & Dragons is dealing with dice and they're doing with pen and paper so they can't use huge numbers so they developed this system called Takko which means to hit armor class 0 and basically what it comes down to if you're not familiar with dungeons dragons is you can only hit enemies that are within a couple levels of you so that means that their damage is actually a much flatter curve and when you go

back to lower level content it still can be engaging me because it's in in large numbers even low level monsters can be a threat to you another way of looking at this is sort of as a giant equation that you're balancing so if you're doing a combat game maybe your programmers are going to give you the the kind of obvious parameters of like damage per shot and then you're gonna say okay well the damage per shot time the times the shots per second is the damage per second then you're gonna take the enemy health you're gonna divide it by the damage per second that's going to give you the time to kill but really the player doesn't know about any of those kind of hidden parameters they know about time to kill so you should be thinking about it as I want this enemy to take the certain amount of time to kill and that means if I want to have a gun that has this many shots per second and that's usually determined by either the the feel or the cadence or the fiction then I'm just

going to multiply our time this sort of unusual parameter of shots to kill the shot skill can be derived in any way that you want but the important part is that you're you're setting the non abstract values that the player is going to really directly experience first and then deriving the rest rather than the other way around a lot of times games are in crunch to the end and designers are making decisions really fast and sort of making compromises based on the tech and so what will happen sometimes is you'll you'll have a genre with a really strong constraint that every game follows but there's really no reason for it to follow so one of the innovations of StarCraft was it took when everybody had done before which was symmetrical factions I mean Warcraft instead it was or his humans but really it's like works versus his pink orcs and and in it and blew that up and it came up with these

asymmetrical factions which you know we're much more difficult to balance but it also means that going forward you can play the game for a lot longer because it's not gonna just to kind of settle into a static state myth is another example where resource management was just bound up in the RTS genre until until we realized that you know don't have to have that part instead of dividing your attention between comma and economy what if we just had the cool combat part we just throughout the economy that led to a game that had a lot more immediacy and felt a lot more like a like a battle simulator rather than a in a strategy game so a few years ago I sat down to play star control 2 which I didn't play when it came out and I was looking at the system requirements and it's like needs a 20 megahertz machine and you have to have a 2x cd-rom

drive and I'm like oh this is great because you know I'm gonna blow the doors off these system requirements would be fine so I started up and I built my ship and I flew out into space it said oh you know we're about to have combat and like I did great and it the combat came up and it was over turns out that there's also a maximum system requirement for this game the clock speed was running way too fast I could never possibly react fast enough to win anything and our brains are sort of similar you know the three pounds of meat they run on on very low wattage their firmware was written millions of years ago and so you shouldn't expect too much out of out of your brain or out of your players brains and one of the ways that games are interesting is they kind of stretch our capacity everyone's called like attention or focus but in this in this circumstance I'm going to call it a capacity so to go back to my Tsar control example like the faster the game was going the the more it stretches my capacity and if he would have been

going at the original slow speed I would have been fine but because it was blowing out the the processor I didn't have enough capacity to actually pull it off so some other examples of this in fact the best example in my opinion is Tetris so Tetris starts out it's actually very simple inputs that you're doing it's very simple mechanic that you're taking advantage of and it starts out very slow right and then it gets pretty quickly up to the point where you're fully engaged and and and it's kind of filling your capacity and then it stays there for quite a long time till like level 16 at least that's my experience when it passes your capacity is too fast and it starts to stack up pretty quickly and you lose when Tetris first came out for the iOS the input is a lot more complicated is a lot more difficult so what they did was they slowed down the pieces a lot but the end result was that you'd be bored for a long time because it was super easy and then all of a sudden it would just clip over it would go just a little bit faster and you

couldn't keep up anymore and you'd lose almost immediately so they rerelease Tetris recently and I strongly suggest you pick it up if you love Tetris what they did was they they removed all the inputs all you're doing now is you're selecting from one of the in this case five places where you wanted to go the answer is five in case you don't know how to play Tetris and and by reducing the complexity they actually increase the speed at which things can happen and this version feels more like the Game Boy Tetris than the original one that they try to do which which shows that it's not just speed that is gaining your capacity is also complexity right and this isn't just input complexities is like complexity and the concepts that you're dealing with or the mechanical complexity of having to do the inputs on a controller or even the information complexity like how much is the game throwing at you they gotta process when I was a kid that they're these two arcade games came out

in 1982 which meant by the time I was old enough to play them they were old but that also meant that nobody was at the machines and I actually got to get up there first was devious had a very simple screen layout it's a 2d shoot-'em-up simple graphic so it was really easy to tell where your ship was really easy to tell where the enemy bullets are coming from and so it was fairly fast and it was a lot of fun on the other side of there Kate was zaxxon which is isometric the backgrounds are really complicated you're having to do with Heights above the terrain and dodging stuff in the environment while these bullets are blending into the background and so it was equally fast but because it was so much more complicated it became very frustrating at least to my ten-year-old self so when you're looking at your game what you need to be doing is thinking about system requirements that you're putting on your players is your game fast and simple like Gradius or is it slow and complicated like civilization or is it

somewhere in between but if you make a game that's fast and complex what you're really making is like flying a jet which is much too hard for for most people there's only a certain number of people in the whole world can even do it if you go on the other side you're not making a game in that direction either because what you're doing is so slow and so simple that it's there's no engagement at all so what you want to do is stay in the I'm gonna call it the game band where you're balancing the speed and the complexity of your game Max Payne is a great example of the this sort of a flexible situation most of the game is fast and the shooting is relatively simple your Freddy one or two guys but sometimes they throw you into a situation where you're fighting a lot of people and then they give you the ability to slow the game down in order to cope with a much more complicated situation horde mode you know not only in Gears but I think this applies to virtually every survivor mode has a kind

of a narrower band that it's allowed to operate in because the setup is generally pretty simple on these games are usually action games they're not about setting up a giant a tower defense system or something and then when the wave comes in and it's very complicated and dealing with a lot of stuff and it can go back very quickly but a smaller game works must die came out and it actually has a very complicated setup you're putting traps you're following guys through different parts of the level and so when the waves come that part's actually just very simple it's just your traps are killing most of the guys and you're just spamming as much of your weapon as possible system requirements come in different types right like there was the processor speed and then there was the CD Drive speed in the earlier example one of the system requirements for games is continuous interactions so almost always

these are things like moving aiming things that you're always doing they're analog and they take up this kind of fast but simple corner of the graph they require a like high dexterity so you usually you'll see one or two of these at a time if it's on the thumb stick it's almost inevitably in in this category the mouse definitely qualifies I'd say the keyboard because it's not it's not continuous maybe doesn't doesn't fit in which is why you know aiming is better on a PC but moving is actually in my opinion better on a console so geometry wars is an excellent example of using up all of somebody's capacity for these continuous actions they there you're always deflecting both of the thumb sticks and they have no other I mean if this guy bombs there's no other inputs in the game so it still stays in that game band it doesn't get too complicated too fast Caleb on the other hand is actually

maybe less accessible to people because you're always using both sticks but then there's also a lot of other stuff going on and so we found that we first introduced halo a lot of people couldn't do two thumb sticks at once and only as the series went on did we see more and more people able to even play the game Killswitch if you don't know is or the kicked off the whole cover shooter a trend that's that we're still seeing today and in some ways this was an improvement over Halo because you're either moving between cover or you're in covering you're shooting but either way you're only using one thumb at a time and that's why I feel like cover shooters are much more accessible so if you're trying to make a game that goes beyond that kind of hardcore shooter market not only making a cover shooter but really separating the two kinds of men think we'll help a lot common actions

sort of take this mid mid range of the graph this is things that you're you're doing but they're usually like simple actions they're not using analog so firing weapons using a melee attack and this requires a lot of attention from your player so you shouldn't have more than two or three of these at a time I think that's why most games don't fill up the face buttons with stuff that you're doing all the time and why a controller with like 18 buttons on it wouldn't actually be an improvement because you just couldn't fill them with you can fill them with more actions than your players have capacity infamous to sort of is infamous for having a lot of things that you can do all on the controller at one time so actually hitting l1 and going into a mode took

the whole mapping of actions that you could do and remapped it something else and while this allowed for a lot of freedom and a lot of expression it also inhibited players when they're playing the game because they were constantly switching modes and so I was really excited to go to Sucker Punch because I loved infamous games and now I get to help them sort of reduce the the complexity of the control scheme a little bit using something that I call random access game design because I love to invent buzzwords but basically that means everything that you can do is always available all the time one action for per button there's no two-step process to cause any of these common actions and what that means is if it doesn't feel in the controller you're gonna have to cut it and that will help your game even though it feels like what you're doing is removing things that the player can do so moving a little bit further damage to the right we get kind of uncommon

actions these are actions that don't happen very often they're very early they're not as often and kind of like under a time pressure so this would be things like jumping or reloading and what's important there is that the player have a kind of a high understanding for what's going on and what their abilities are because you have more than you can really keep in mind at any given time the the game needs to be kind of cluing you in that now's a good time to use it that's what she said is sort of an example of an uncommon action you can use to be participating in conversation you're trying to like communicate and then somebody will say something that that's what's easy simple just pop in your mind right and we're really thinking about it constantly it was an uncommon thing but as soon as the moment is right you know it's time to use it so bold storm has a great example of this sort of that's what she said kind of action in its skill shots so skill

starts a super important in the game they they drive your progression they're heavily featured in the marketing and one kind of skill shot is the environmental skill shot where you're knocking some guy into a spike trap or off the cliff or into a oven or lots of disturbing ways to kill people and it's important that for these uncommon actions that the player is not in control of the pace so the onus is not on him to control when this is going to happen it's just sort of the environment dictates when the these options are available and that means there's these really strong external cues like if there's no giant spike trap then you can't use your spike trap kill shot on the other hand the weapons specific skill shots don't follow follow these rules they are set by the player you can theoretically use them at anything at any time but when you're playing yeah you actually end up kind of going into the menu and figuring out which one

you're gonna try to do next then going back into the game it's not a very fluid process and usually these kill shots are not as efficient as regular just shooting guys and so you're not as incentive to do them and then finally in all the way in the lower right are prompted actions so these are actions that are totally on there's no time pressure and that means you can have as many of these as you want because you're gonna have to tell the player how to do them every single time Borderlands is a great example of this they kind of took the action gameplay that gearbox had been doing before and they added this off offline almost loot stream where you can gather money after a fight you're comparing weapons the UI you're equipping items but none of this has to be done in the context of actions so it's it's just it's just addition it's not actually increasing the system requirements for

the game except when it comes to health packs health packs are picked up in the exact same way but a lot of times you're in a fight and you need it right now and so you're trying to look around for the one health pack and you're trying to aim at it so you can pick it up this puts a lot of pressure into into a corner of the graph that doesn't support pressure very well so moving past constraints constraints allow you to limit the area that the player is going to be able to interact and so the next part of the design process is to look at the interactions let's talk about pop man so even at a really early age and I think this might have been my very first video game I was super sensitive to details and because of that I never really got into pac-man and I didn't find out until 30 years later why basically in Japan they use 4-way stakes right and that's why the cardinal directions of pac-man make total sense and it works really

well but American arcades use eight-way sticks and a lot of times they didn't ship over the whole arcade unit they just shipped over the ROM so they were plugging into an eight-way arcade stick and so pac-man was all of a sudden getting these diagonal inputs which would it would ignore just fine except for it also ignores the next input so if you're trying to roll around a corner in in pac-man in an American arcade you're gonna just ram into the wall and all Ellison pac-man is gonna stop you're gonna get caught by the ghosts another example of how physical limitations impact your interactions is Dungeon Hunter it's a iOS game very similar to like a Diablo and like a ton of iOS games it has virtual buttons which means your fingers are always on the screen and you're trying to do reaction times test reaction-time tests on something that's not giving you any feedback it's very difficult not to say it's not a good game it actually is a pretty good game but the you really feel

the limitations oops whereas Infinity Blade uses gestures which is what the device is meant for it has pretty infrequent input compared to the length of a fight so most of the time you're looking at the fantastic graphics and even though it's a game about timing it's not reaction time test you can see every blow being Telegraph it's more about timing you're blocked or dodge to the animation and not reacting to what's on the screen and these guys did an excellent job of porting the details of a punch out as well responsiveness is a big part of interactions doing something and then having something happen gun stringer is got a ton of personality and you're basically moving a crosshair with your with your hand in front of the Kinect but it's heavily smooth because they didn't want a lot of jittery on the crosshair and so what that means is that

you're you're experiencing about a hundred milliseconds of delay between moving your hand and seeing what's happening on the screen and your brain actually will take events that are less than 100 milliseconds apart and merge them into one event that's why I like sound sync and a movie works even though it's not not exact but on the other hand if it's more than 100 milliseconds you're gonna perceive it as two different things and so if it's intended to be one thing it's gonna be very disorienting fruit ninja on the other hand kind of smooths the spline in real time when you're swing your hand across the screen there's no no delay and so it feels a lot more one-to-one responsiveness is a is a huge part of a new shooter game Halo 3 clocked in at about 100 milliseconds which was just a little over 3 frames at 30 frames per second which is again right right at that line for things feeling simultaneous Killzone 2 on the other

hand because of how they were doing their rendering had up to 150 milliseconds of kind of delay before your input was recognized and that turned off a lot of like professional gamers who were sensitive enough to need that instant responsiveness and right now that the current king of this is the collagen you guys they they run at 60 and they have all I mean they're approaching kind of the theoretical minimum for how much responsiveness is lay you're gonna have even after you've got the players input you have to interpret it and we were making halo time splitters it was a game we were playing a lot and they were practicing auto-aim which meant you know your bullets are coming out of the gun and kind of getting aimed toward enemies and because of that there were all kinds of frustrations where you were trying to do one thing and you couldn't like your gun would actually be behind a wall even though your crosshair wasn't so you'd be

shooting the wall and not realize it or you'd be trying to turn and because they didn't want you to turn it too fast you just you just couldn't get around and some guy was shooting you in the back so for halo we went more in an interpretive direction with aim assist where we're trying to figure out what the players doing instead of what they're saying they're doing that means that in Halo you shoot from the crosshair not from the gun you can stick the gun in a bush and your your bullets will come right out of the center of the camera pegging the stick actually makes you turn faster like it continued to accelerate because essentially what you're telling us is man I wish the stick went really far and then when you stop giving us input over an enemy it actually slows down on them right away much faster than if you if you slow down just over an object that you're not probably intending to shoot at it's like Cooper does a super good job of this as well when you when you jump in sly it's it's mostly vertical so if

you do something you didn't intend you can probably get back to where you were it steers you the entire time toward where you're gonna land and that allows you to really stick the landing a lot of cases where as prince of persia does more of a job prediction thing where they're trying to figure out where you're jumping to at the moment that you jump and it's a mostly horizontal jump so if you tell it to jump and it can't find a place for you to go it's just gonna throw you out in this space which is fine because then you can rewind it and but a lot of times you felt like you you you weren't in control of what the prince was doing I'll just skip that slide another another way that interactions kind of are flavored in one detail can affect the entire game you can see in Resident Evil where they're trying to get you aim under pressure they're trying to have scarce resources but what that ended up being in their game with the details they picked was you just can't move while you're shooting which is a high pressure but

also feels like you're limited and then they just greatly restrict the ammo so anytime you miss yeah you feel like you've failed whereas deads based is a much better job this because they are they're requiring you to shoot at different points and those points are often moving a lot more than ahead so you're still under an equal amount of pressure but they can give you really responsive controls because what they're asking you to do is much more difficult and then they do a good job of algorithmically figuring out what you need what ammo you're using and so you never feel like you have enough ammo but secretly you always do rewards are a big impact in interactions and what's often difficult is to line up the rewards with a certain kind of activity so you don't want them to clash in lots and lots of games that have good or evil systems they're promising you this choice and then they're going to give you power if you make it to the end of either the good or evil spectrum which

means you really have a choice and so the rewards are diminishing the activity anyway I think of it like this calories are like a fundamental human need we've evolved all these systems to reinforce the many calories and so what happens is when you eat sugar your body releases insulin before you get the calories kind of in anticipation you're gonna you're in a proc process the glucose loses sugar and you're gonna get those calories right and then it releases serotonin which is sort of a hormone that that makes you feel satisfied the problem comes when you when you eat fake sugar your body still releases insulin but there's no glucose to process you don't actually get the calories so no serotonin is released and you end up with like hypoglycemia low blood sugar or insulin problems and so fake sugar actually in a lot of cases leads to weight gain because you're what you really want is serotonin and not just calories games are very similar

I believe that players have sort of fundamental of psychological needs it doesn't matter how you classify them you know they can be subdivided in lots of different ways and when you're playing a game the game kind of sets up the conditions for success and makes you want to aspire to them and so you're making these attempts and eventually you're going to succeed and then only after you experience actual success is dopamine gonna get released and you get that kind of feeling of satisfaction so just like with calories the the real reward their chemical reward that you're seeking can only happen after a entick success whereas gamification is actually frustrating in a similar way to fake sugar because it skips the actual success and so instead of getting dopamine you're just gonna get frustrated and you're gonna feel like this activity is futile even though it looks from the outside just like a regular success cycle you're cutting out the actual important part so there's

lots of ways to increase this this the effectiveness of this cycle what you want to do is increase the connection between the success and the dopamine by making sure that the reward the player matched the activity that they were participating in both I'm kind of the the flavor or the spirit of the activity the timing of the activity and you don't want to undercut all the the failed attempts that the player made by making the success too easy or unavoidable Assassin's Creed has struggled with this is it you know it's much better than the most recent games the first one they give you things to do like flag collecting where they forced you to do it when you picked up one they were like great you only have a million more to go and they didn't really even tell you what the reward was gonna be and from the outside that that looks like a reward loop but because of the activity doesn't match the reward in any way it's not very effective at getting

players to keep playing the investigations were really similar to this where they were forced you to do some of it and then there was this really unclear all your assassinations are gonna get easier but they were mistaking a character defect completion ISM with an actual motivation and there there were conflating just like checking off boxes with actually successfully meeting some needs Batman on the other hand nails this really really well like the interview tapes they never tell you that they're in there you just kind of find one and it opens up the UI so it's not forced they're rewarding your autonomy really well and then the reward that you get is instant the first time you pick one up you get a little bit of story and then you're like okay well I'm gonna I'm gonna pick it up pick up another one I'm gonna get a little bit more story and the trophies kind of do this in a similar way for a different need you again you find them by exploring they were forced on you and

they're rewarding your competence of finding these things with something that a competent player really values was his trophies I looked it up in like 20% of all the trophies in Batman are about ruler trophies and then also the Riddler like gives you a voicemail or something where it's really pissed and that is something that a competent player really is attracted to another way you can increase the power of this cycle is by just amplifying the amount of dopamine that the player gets if you want a good lesson in this you should just go to Las Vegas because those guys are dopamine masters game center which is Apple's sort of a platform achievement system it's terrible at this the the the notifications that you got an achievement or are delayed by seemingly arbitrary ways the the skin is like this

antique billiard table which this does not match at all the feeling that you're getting from inside of the game there's no celebration or sort of pomp to to the awards at all but they're not actually valuable which means that they're poor at giving you dopamine Batman on the other hand does this grace the rewards are always given right at the at the moment where you realize that you were successful it's not given at the moment that you're actually successful it actually delays it a little bit so you get this chance to anticipate and go oh yeah I know it's coming then there's this there's bats everywhere and the music swells and then like in the case of the column you actually like count out every single point and this meter goes up and it takes a long time and then what they're giving you is actually valuable you can actually use those points to purchase something so you attach more meaning to them and so that makes it great for dopamine and then find a way to to strengthen this reward

cycle is to reward an activity with more of that activity this is probably the most important thing I ever learned GDC was after a talk I gave I was talking to Scott Rigby he's a psychologist he's researcher and he also does like consulting for different game companies I said what so what is the best reward for for when somebody does something and he said exactly this warden activity with more of that activity so the best example in my mind this is the original World of Warcraft there's there's all kinds of things you can do in the game leveling gathering resources raiding playing PvP and all of those lead to more of the same things so as you level they give you more levels when you're crafting that allows you to craft new things you're pvping that give you items that are specifically geared to PvP and this sort of one outlier which is you have to level all the way through before you can really get into the rating and because every designer has to have a personal theory about why wow is

this will failure here's mine the current Wow they've sort of lost that flow where sometimes you know they sped up the leveling you're gonna hit the level cap a lot earlier and if you're not a good crafter then you can't go rating because you don't have the right craft your bonuses and then there's all kinds of things that kind of go back and is it's just a mess where you can't you end up having to do things that you don't want to do and you know in order to get the rewards that you want you've got all kinds of like cross purposes and the different reward cycles so yeah again totally dismal failure Batman on the other hand is really well one of them that I talked about earlier is comma XP so you're you're building these combos you're you're fighting well and what that does is that gives you the ability to buy upgrades that make you fight better so instantly then your incentive to go back into the fight and

and continue this activity you actually enjoy the challenge rooms is it I regained a great example of this you know you beat a series of challenge rooms and that unlocks more challenge rooms so the next step in the design process release my design process beyond interactions it's something that I'm calling Harmonix and this is still something that I'm I'm kind of figuring out as we go so this is gonna feel a little bit more experimental than some of the other sections so I start with a concept like a 2d platformer right Mario Brothers your guys weren't into the right sometimes you have to kill enemy something to jump pits and then ultimately you're gonna you're gonna fight a boss at the end and there's lots of different games that fall into that sort of same rough concept but and then they do the details different you know I'll explore a little bit how I think that effects the ultimate game so the constraints in in Mario you always have to go to the right you can jump you can stomp on guys and then you're gonna get mushrooms that

will make you a big which is essentially like one hit point and what that leads to is sort of one frequency of a challenge where you'll run to the right and then you'll jump a bit then you'll run to the right then you jump it or you'll stop on an enemy or you jump another pit and then the health sort of system leads to just just two levels right you're small you're big you're small you're big you fall in a pit you're dead sorry Hedgehog on the other hand it still go right but instead of jumping and stomping on guys you actually just jump into guys so that helps them maintain this sense of momentum and and forward progress and then you collect rings and much like being big in Mario it's it's kind of a life insurance policy you should have to get hit you're gonna lose all your rings like you're not gonna die so you the-the-the the cycle is a little

bit less interesting at first because instead of jump over a pit or a stop on an enemy you jump over a pit and jump into an enemy so it's a little bit more of a simple action and then the health system basically as soon as you can hit your ulsan you're switching mode you're scrambling to correct collect rings but it's really easy to pick up a ring after you get hit and so you don't actually spend very much time in that collecting ring mentality you almost always just have rings Megaman on the other hand adds a lot more constraints that make their frequencies and harmonics a little bit more interesting so you're not always going right sometimes you're going up sometimes you're going down Megaman has hit points and and the enemies have hit points right so instead of a single moment of interaction sometimes you're taking damage sometimes they're taking damage you get these small little oscillations

and then because you have health bar it'll start out full and I slowly go down slow go down and so you finish a level or get to power up and go back up same thing happens with amo so what's interesting about harmonics is that the same constraints in a different circumstance can produce just drastically different harmonics and for a platformer that a change of keys if you will is a boss fight so in mario basically the harmonics just flatten right if you're big you run right through Bowser and you win and if you're small you're gonna get stopped by Bowser and maybe you can make it through but it's mostly luck so the the constraints of Mario actually don't don't don't hold up another bat for the boss fight conditions sonic holds up maybe a little bit better but the problem with the boss fight in Sonic is that the enemy boss is going to hurt you

by running into you and you're gonna hurt the boss by running into him and so determining who's running into who it just becomes kind of a fuzzy concept for the player and also you're gonna jump into a boss and whereas normal enemies who just die instantly the boss is not going to die instantly so I'll listen you're gonna be standing in the boss and maybe he's just going to hit you before you can get out of them and then the the Ring concept breaks down pretty pretty heavily in a boss fight because you basically have infinite life as long as you have one ring and you can always grab before you can hit again Mega Man on the other hand with a little bit more complexity in their constraints they actually there's some blossoms when you get to a boss so the health bar is set up a longer fight between you and a boss and the fact that you can have different ammo types set up this whole kind of matrix of what weapon you're going to use against you so some other things that control harmonics are our pace setters not on purpose both of the talks

I've given before we're about pace setters so increasing the the time between shots for the sniper rifle will slow down the harmonics and decreasing the the rate of velocity of the plasma bolt actually makes the fight shorter so sort of before I had really landed on this concept I'm harmonics it was already really heavy in my mind but but another thing that probably has more info influence on harmonics is things that reset the cycle so like the end of a turn in magic all all of your cards on tap everything's reset and it's something the player controls so it becomes a game mechanic in and of itself whereas dying in BioShock is not under the players control and because you respawn in the video chamber with all your health but not the ammo that you expended you get these weird kind of out of syncs between your the amount of health that you have and the amount of

ammo that you have and kind of the state of the world and and and their harmonics kind of diverge so just look take another look at kind of how harmonics work you know what I'm just gonna I'm just this is the last four slides and I don't think I'm gonna make it but basically the four games that I've been playing most the last year Super Meat Boy legendary halo dead space and dark souls all have extremely similar harmonics even though the mechanics are utterly different so I would encourage you to play all of those games with a mind toward how are these games the same and that will give you a little bit more insight into how mechanics will create the overall feel of your game and then that and will have just a very very

short time for questions [Applause] yeah you have to walk to one of the mics if you want to ask questions which I start calling on people sorry yeah I'd rather talk about Dark Souls I actually I put in like 120 hours in Dark Souls with no spoilers or hints and it is it is a masterpiece of of game design because every decision that they make because the stakes are so high just

impacts the player almost directly so if you haven't played Dark Souls I would strongly encourage that so I didn't really grasp from your slides specific examples of what harmonics are in a one or two sentence thing could you expound a little bit yeah so you can strain the player right and they have certain interactions that they can participate in and those interactions sort of have a natural frequency like the the shield recharge in Halo is a super example of something that happens at a certain rate and it's a very predictable timing and what will happen is when you take all of those different frequencies and you combine them they're either going to be in sync and being in kind of a good in tune I suppose or they're going to be at odds with each other and a lot of times games that feel solid and like

well-constructed that they have a lot of synchronicity between the harmonics whereas a game that feels off somehow a lot of times what that means is that the frequencies are a little bit separate what you're really saying is that good harmonics are games that are well balanced no no balance at least to me now it's been timing and system balance yeah they have a they have a good rhythm to them yeah and I'm when I when I first started I very much looked at game design as sort of an engineering problem like come up with the right mechanics that will build the right system that will produce the right experience and as I go on I see it more as a almost composing right coming up with melodies that repeat and sort of a predictable flow across across the game yeah so it's very clear that you have a

very rigorous and analytical approach to games what I'm interested to hear about is I mean it's easy or not easy but so you're looking at games that have already been finished and analyzing the harmonics how are these different mechanisms interacting with each other can you speak towards looking at those mechanisms but from inception so from from creating a game is there a way to anticipate or like how would you go about applying these same types of mechanics and mentalities to building a system that interacts well together yeah I think um go listen to a really good trance song right they're gonna start with it with a with a beat and then they're gonna layer on lots of different harmonies then at some point they're gonna change it all up and I think you can approach game design sort of a similar way you figure out what the what is this what is the pacesetter for your game what are the things are going to start a new kind of familiar theme and

then make sure that the the the timing for those are all gonna all gonna line up and then you don't have a game where everything happens at the same frequency because then you're gonna get the same experience over and over and use different frequencies that still line up to produce a lot of variety in the gameplay and this is obviously something that I'm you know I'm going in the future I'm gonna try to do a little bit more intentionally from the start sort of retro actively okay if there's no more questions thanks for coming get to the airport get on your planes go back to all your various places you