Crashlands: Design by Chaos

[Music] all right here buddy my name is Seth castor and I am the game programmer and co-founder of the studio butterscotch shenanigans a little bit about our studio we were founded in November of 2012 we started as a two-person team and we've since expanded to a four-person team so we've just grown massively we've launched five titles since our inception and the most notable title we've launched his crash-lands which is an action crafting RPG we launched it in January of 2016 and in that year it received numerous mobile game of the Year awards sold over half a million units and received a 93 Metacritic score for mobile so by all accounts it's okay and people seem to like it and before I go into the meat of this talk I need to remind you guys that we only have 20 minutes which means what I'm about to say has absolutely no nuance whatsoever so if you have a bunch of questions at the end of this that makes a lot of sense and you're gonna have to just think about it I guess all right so because crash-lands did

pretty well for us we get this question a lot which is how do we go about designing our games and to answer that question I'm gonna ask a few questions of my own why do studios crunch why do studios go bankrupt and why do projects can cancel well these are the kinds of things that you would expect to see in a studio that has plans and so we've managed to sidestep a lot of these problems or mitigate them entirely by never having a plan of any kind whatsoever for example we don't have deadlines so we have no reason to crunch and deadlines are the kinds of things you have when you have plants and this kind of anti planning philosophy goes all the way down into our game development style because to answer the question how do we design our games well we don't we just iterate we've never done a game design document we don't really do anything like what you would call concept art and if you were to pick any one thing out of any

one of our games we didn't know that that was gonna be in there until the moment we put it in so for those of you who are more customs to a much more structured game development style you're probably thinking okay how does that work that sounds like absolute chaos so that's what we're gonna talk about the first thing to know about iterative gaming development is that it's not random the iteration is directional and the thing that you're always iterating towards is the vision of the game which is just the core idea we know a lot of developers they get really fixated on trying to find the perfect game idea and they bounce from project to project just trying to find that that one gem at our studio we believe that there is no such thing as a good game idea or a bad game idea there's just good or bad if execution of a game idea and we believe this because we've seen ideas that were just so far out there like be a goat and break things that have turned into huge classics that everybody loves that we just come to

believe that you can just do whatever you want and as long as you understand the fundamentals of what makes your game good and follow that path then it just doesn't really matter so so this is the core idea that we have as a studio is just what your game idea is is completely irrelevant so if that's the case why do you really need a game idea it's because your studio is a boat and it has limited resources on it and at some point you got to get to land and restock right in other words you got to launch a game and if you hand everybody on your team a paddle and they all start rowing in different directions you're just gonna spin in place and everybody's gonna get scurvy so so the point of a vision is just to get everybody paddling all in the same direction and that's really all that it's for so in our studio the way that we go about sort of structuring our vision development process if you could call it that is we keep it very simple first we keep it brief so single phrase or sentence once

you get into paragraph territory that starts to look like a game design document which is a plan and once you have one of those you'd die so don't do that we also keep it very high-level so we don't really go into a lot of specifics about the story of the characters any of that stuff we just focus on what we would say our mechanics experiences or feelings and the reason we focus on these is because these are simple touchstones that we can always refer back to and ask is the game feeling like this or do we have this mechanic and are we executing it really well so to give you an example of how we did this with crash lands the crash lands vision is it's a light-hearted open world crafting adventure with tamable creatures and lots of jokes we ended with a lot more than that and we also at this time we didn't know who the character was what their motivations were where the story was where the game was any of that stuff these were just the targets that we were trying to hit and then we got started so once you have your vision it's time to start prototyping the prototype only

exists as a as a sort of anchor point for your iteration process and that's really all it is this is the thing that will become your game in the end so to start your prototype off on the right foot you're going to want to find the foundational piece so if your vision has something that you believe to be sort of the core element that if you nail that then everything else will kind of spin off of that but that's where you want to focus your efforts for that early prototyping phase crash-lands was a crafting game and so we thought if we get the crafting then we're good so you can't craft unless you're collecting stuff and so the first thing we had to do was focus around that acquiring things in the world and then building stuff out of them so with that in mind we started prototyping and on day one we had this which is a gray rect or gray circle that picks up hand placed leaves there's only ten leaves the camera doesn't move and once you pick up all ten leaves they craft into a pair of

leaf shoes I guess so that is our crash-lands prototype on day one and now we can start going through the iteration process so we do in-house for iteration process is first we play the game and we try to play the game in a very sort of you got to say your ego aside and you want to think about what would a player perceive of this if I just handed this to a random person so you think about it as if somebody else made the game and then you want to evaluate how is this current state of the game lining up to the vision that we originally put forth and where are we just really missing the mark and what kinds of systems do we need to shore up you can go ahead and invent solutions for for those problems bring those back to the team come to a consensus if it's something huge like if you say we need to rip out a system entirely so you go ahead and do that you implement those things and then the loop starts over again and so the important thing to know about this iteration concept is that a lot of teams start to

forecast at this point once they start working on a system they say well once we get to combat in we're gonna need to buff debuff system and once we have that we're gonna need combos or whatever and you start forecasting what the game's gonna feel like three five ten twenty iteration cycles down the road well guess what that is was planning so don't do that you should iterate based on what the game is not what it will be because all that time you spent forecasting you're just gonna be wrong because you can't know what the game's gonna be like so with that in mind let's take a look at what crash-lands evolved into over time from this weird gray circle picking up leaves into what it is today so here we were on day one after a couple weeks literally everything's different all the art has changed and this is all placeholder art as well because the art we're just using to sort of try different color palettes and shapes and stuff to kind of see what's feeling good is this is kind of hitting the target that we set forth in the vision with the

full intention of doing a repast on these assets later at this point we also gave the character the ability to remove its arms and replace them with anything in the environment was an insane and terrible idea and so that didn't last more than this week but this is the the early phase of the game where it all bets are off anything's on the table we can do whatever we want and just see how does this line up with the vision after a couple months the game starts to solidify and look a lot more like the modern version of the game today and we have we have one biome in place and it's not fully fleshed out we're learning about how to make a scalable and robust crafting system we have lots and lots of stuff in the world we're starting to figure out who the character is but we don't care that much yet and you can craft yourself armor and weapons and there's combat and all that stuff so the first thing we had to do was give that crafting system in and once we have that then we can start kind of layering the rest on top and then finally as we approach launch this is what the game

turned into we ended up with about a sixty thousand word story and yes you can put story into your game through this weird chaotic process and we'll talk about that a little bit we have nearly a thousand unique items in the game 500 craftable recipes just tons and tons of content none of which we knew was going to be there and until it was so what we like to think about this process as the game discovering itself over time because as as iterators rather than designers it's our job to simply play the game listen to listen to it feel it and kind of kind of feel out what it seems to want to become and just follow the trails of what's fun if something is boring we try to fix it or if we can't fix it we get rid of it like removing your arms and replacing them with weird things and over time you can see that the game solidifies very quickly because the first three screenshots are the closest together in time but the last two screenshots look the most similar because those early

days the game is kind of like a like a white-hot ball of magma you know and you kids just very liquid and fluid but the longer it sits the more it solidifies the more it cools and the more it just sticks to whatever shape you give it so so that is the iterative process that we use to make our games with no planning at all but there are some things that if you want to do this in your studio that if you take these philosophies into your workflow then it'll it'll help you kind of work through these these iteration cycles much more quickly so the first thing that we do is we focus really hard on speed because iteration is a compounding process if you can get 20 iteration cycles in in the same time that somebody else can get three you're just gonna have a better game in the same amount of time it's gonna be much more cohesive and you're gonna understand it much better because of that our studio has a philosophy called take the time to go fast which is it's just a little saying where you know if if you think there's a tool or something you can learn or something that you need

that if that's gonna save 20 minutes a day for you on something but it takes you eight hours to figure it out it's still worth it because that's a permanent gain in your speed right and so so we always focus on speed we do lots and lots of game jams to sort of test ourselves and we'll make games in just a couple of hours just to see what we're capable of and try to find our weaknesses in our processes we also love the phrase can we not which is just a way of asking yourself am i doing stuff that's a waste of time and what would I what would happen if I just didn't do those things because what we found is that speed of development largely comes from cutting out things that you just don't need to be doing whereas most people tend to try to focus on just doing the same things faster and that's where you start crunching that's where you get burnout that's where you get all these other problems so because of this we actually only have to work four days a week and it's still fine we also adhere pretty strongly to the Goodenough principle which can be summarized by the phrase perfect is the enemy of good and

to kind of give you an example of how we do this when we were working on crash lands early on we needed to get the first creature into the game and we had a set of criteria on what this creature had to do for starters needed a simple avoidable attack because the first creature that the player is going to encounter and so they need to be able to understand how to not get killed by this thing it should be pretty friendly looking even though it's dangerous but we don't want to scare players away because again it's a light-hearted game it needs to fit with the aesthetic of the environment which is easy for us because we only have one artist but for other teams this may be a little bit more of a challenge I needs to have very few moving parts because we don't know how creatures work yet and we don't know how we're gonna animate them and we don't have a plan for that so we're just gonna make something that's easy to implement so we handed this list to our artists in the morning and by noon he had created the wom pit which is a one legged creature so few moving parts no problem there and it

stomps on things which is a simple attack you can see it go through the air and land so it checks all the boxes so we said is this good enough yes it is we put it in and we had it fully implemented by the end of that day and that's the same creature that's still the first creature players encounter and crash-lands the reason that we stick to this this strong good enough principle is that as a developer you have to understand what do players actually care about because when a player encounters this WOM pit their reaction is gonna be help I'm being chased by a one-legged hippo that's sort of where their thought process ends because it's a threat to them it's coming after them and once it checks all those boxes they know they don't care anymore so a player isn't going to sort of behave the same way that maybe like a video game producer looking for perfection might the players not going to lean in and say you know the ears on that whomp it could be slightly larger I think they don't care because however

you present the world as long as what you've created is good enough and fits the player stops asking questions at that point so if you're having big debates about whether your main character should be wearing jeans or khakis maybe just don't another thing that we like to use is using prototypes to crush debates this phrase will save you so much time in your development cycle that it's unbelievable how long would it take to prototype this idea when we were working on crash-lands we had these floors the players could build and they could sort of lay them out on the ground but players weren't building them and they weren't using them for anything and it turned out that floors were basically a useless item that we hadn't given any any purpose so Sam our artist came up with an idea on what we could do to make floor is useful Adam and I the other two members of the team didn't understand how his idea would work or how it'd be feasible or even why it would be good and so of course we did the natural thing and we argued about it for three hours after three hours we

didn't come to a resolution we walked away in a huff i sat down at my computer and I prototyped it and seven minutes turned out it was a great idea and that's what we have in the game today so if you find yourself arguing for a long time just ask how long would it take to prototype so the final piece of this iterative design philosophy is that a lot of people get really fixated on story and the question is can you get stories into your game using this sort of chaotic design method so to answer that I'm a total story of my own in the beginning of crash-lands development we got combat in the first couple months and the player was taking damage but couldn't heal and so so I said to Sam our artist I said we need an item the player can just pick up off the ground and eat it to heal and so he sent me this flower and I'm sitting there thinking who eats flowers this is weird

so I just gave it a name and called it bacon weed and now it heals you and it makes perfect sense later on in the development cycle we we added farming where the player could sort of plant stuff and put things in pots and regrow stuff and so potted bacon weed was born and as the programmer I was also writing the tooltips for the items as we implemented them into the game and is my job to write the tooltip for potted bacon weed this was the 30th thing I'd written a tooltip for that day I didn't know what to say cuz it's making weed in a pot I mean what more do you need so I just made up some nonsense and I wrote legend says if you leave a potted bacon weed by your back step the bacon weed fairy will steal your teeth we said hit and then we moved on about a year later we wanted boss fights in the game and we didn't really know what the bosses should be and so we kind of combed back through all of our tooltips and we kept

coming back to this bacon weed fairy and we thought this has to be a boss fight so we created a huge elaborate quest chain that the players go through and at the end of that quest chain they confront the bacon weed fairy whom they have to defeat to get a healing trinket that's one of the best items in the game and the bacon weed fairy is a crazy bullet-hell fight that changes everything about the way players think about the game and it's also one of the coolest moments that we see whenever youtubers or streamers play the game because they're so surprised and taken aback by this as as you would expect so the way that you then go about creating story using this design process is for starters don't worry about it finish the game first because the game is the setting for your story so just as the game takes shape over time you're just gonna be looking around for little pieces of of hints at a story you're just gonna pick them up and weave them together and just kind of let those

stories emerge from this weird thing that you made and above all when you're telling your stories be unapologetic about it just let the game tell the stories that it wants to tell and don't try to force it to be something that it's not because maybe it's Tuesday knee or too weird or whatever because as long as you interpret the game honestly as it as it seems to want to be interpreted you're gonna end up with something that fits really well and is also very interesting to players so that's that's our whole design process so as a quick recap of the main points in case you guys want to take a snapshot of it you can I guess use your phones for that so so for starters ideas don't matter don't worry about it just start making stuff quit worrying so much that's just in general just life advice don't plan everything or anything if you can help it because you can't know the future and you're gonna waste time planning and then waste time being wrong later so don't do that also don't sweat the small stuff also don't sweat the big stuff either

just stop sweating it's not a good look speed is your best friend so focus on that and everything else will fall into place and don't worry about trying to tell stories in your games just find them because they're just laying around so just pick them up alright thank you guys very much appreciate it also we we do have a podcast under studio which we talk about it's kind of stuff all the time so if you'd like check that out we have a link up there you're gonna find me on Twitter you cannot so good luck all right does anybody have questions yeah so great pulp thinks you mentioned in the beginning we were like four people what was bigger afterwards so we started it as two and we are now four okay and do you think it would work for like a much bigger team of twenty or more I simply cannot say I think one of one

of the things that's really important for our design style is autonomy where it's sort of like you know we do things by targets so some general idea needs to happen like with the whomping example we just have our checklist for an artist and we don't care what comes out as long as it meets those requirements and so he made that thing and we put it in without any oversight really or anything like that so I think it can work with a larger team but you would need people who you really trust to be that autonomous but at the same time I think the larger a team gets you know the more people want to know what's happening and the more planning you seem to need so one of the questions people rarely ask is you know how small can we just keep our team or do we need to team this big and what if we split the team and make two games in the same time right so I think I can't say but I've never been that comfortable on large teams anyways so who knows thanks thanks yep yeah so I

just want to thank you for all the hours that I spent playing this game is awesome yeah no problem I wanted to ask about I really fell in love with the creatures and especially that in beginning moment yes can you talk about the design process that led to that yeah so that design process was just this but I'll kind of recap how it went which is so we have we have a moment in crash-lands where you can collect items and then you can create this weird thing for each creature and you can give it to the creature and we do sort of like a Pokemon style you know it flashes on screen what it's evolving or whatever but then at the moment that it's about to evolve we flip the screen into this crazy view with weird words flying across the screen and we turn on some dubstep and then the creature just kind of flies in your face a whole bunch and big words flash on the screen saying creature am bigand and that's and then it just it

just happens for a few seconds then it just goes away where the players kind of going like what what just happened I would literally when I was playing this game on my commute I'd be like oh god I can't watch this on the bus because I wanted to save it for home yeah yeah became one of those special moments and so so where that came from was was actually I had been doing the programming for this and Sam had created some of the art for this and beginning scene and we originally just had some very mundane music and I'm sitting there programming it and I thought this isn't this isn't anything so I kind of snipped up some some skrillex or something i just threw it in there just as sort of a demo and i didn't tell Sam about it and so then he was play testing the game and that he just started screaming so we were like yep that's that's the thing but again that was just kind of again that just came through interpreting the game and just saying this isn't good enough you know this isn't hitting what we needed to hit we need a joke here and so that's where that came from yep hi

I'm wondering about play testing your game and if you guys do any play testing with players outside your studio yeah and specific you talked about the floors and how players weren't using the floors who are the players like if the game wasn't out yet yeah so we do we do a kind of a series of testing moments where at the very beginning we will just do one on one sit downs where we'll bring in somebody who's never seen the game before who isn't afraid to tell us that they hate it and we we don't ever ask them questions we just sit down with them and as they play we just take notes and just watch what they're doing and see if our ideas are actually landing because it's easy as a developer to say here comes the funny part and then they're gonna be like hurt and try to you know like when you show a friend a YouTube video and then there's that painful so we don't do that we just we just observe and so we do that early on in development and then later on we reach

out to our community of players and we do sort of a small alpha where because these are these are sort of our bigger fans and they are very sort of forgiving of if the game crashes or has terrible bugs or whatever so we do that and then we do a much bigger beta later on through through invites so yeah we do have a pretty stepwise testing process that we go through yeah okay so I love iteration I love prototypes I agree that that's how you make your games not suck yeah I said that's awesome what I've noticed happening on development teams that have been part of is that 40 iterations down the road you're like we're gonna implement this feature it's gonna test it today and you're like a week later you're like wow that was a lot harder to implement because there's a whole lot of code in here that's kind of bad yeah and so at what point or if ever do you be like alright we like what we've got tear it down build it the right way so that we can continue prototyping because now our momentum is being sucked yeah so one of the one of

the key ideas that you want to hit when you're going through these iterations is trying to get the current state of the game to be in a really good spot so as opposed to saying you know the game doesn't feel very good but if we just add 10 new systems I feel like that it's gonna do it you know because now you're six months down the road you're gonna get all those systems done and the game is worse now because it has a bunch of systems with no content so one of the things we've noticed happening when we do these iterations is that if we leave a sort of a stone unturned where we leave some kind of a weak special effect or something that just doesn't feel good to use then we will go ahead and forge ahead without that thing and we'll just keep developing the game and leave that poor that poor underused system behind because it just wasn't doing what we needed and so so that's not good right because now you have a system that a player will encounter thinking this is gonna be great and then it's just kind of bland and then they just forget about it so we've really tried to take seriously the idea of whatever you put

in the game nail it so a lot of people try to save polish for last and I think if you do that under the iterative cycle then you're gonna end up just with a lot of sad systems that have never been finished because you didn't want to use them you didn't want to explore them and you didn't want to play with them to do your your comparisons to your vision so you should always try to shore up your systems to be in the best possible state before you think about moving to the next thing good time for yep I wanted to ask why do you use game maker studio I mean I'm a master ace myself for that but I wanted to ask why do it's the best we've tried using lots of other engines for 2d development and we've found that there's we just can't do anything faster than we can do in game maker studio we've actually we've even done a series of jams where every Monday we started a game at 8 o'clock in the morning and had it published on google play by 6:00 p.m. with a full set of loops in-app

purchases the whole thing some of those games even got featured and so with that with that kind of speed since that is our primary concern using this development style we just have to go with whatever tool lets us do what we want to do the fastest and that just that just comes to be game maker studio so so with your iterative process I guess you said you you're trying to just cut out as much plane as you can how do you know when you're getting to that point where you're ready to launch and I guess if you're just always kind of yeah yeah you can always keep going right right yeah it basically just amounts to you for us you just gotta kind of feel it out so sort of a lot of it comes down to comparing the game to the vision right and when you feel like you've just hit everything that you wanted to hit you got to wrap it up and and move on because you will just run out of money at some point or people will just lose patience so if you're trying to build a pipe at the beginning of your of your dev cycle for your fans or whatever and

then it takes six years because you just gotta get it right then you're gonna have a hard time and that's partially where the Goodenough principle comes in as well where you have to be willing to accept that perfection is impossible to attain and once something is is good and good enough that your players will really be into it then you move on to the next thing if you have a lot of ideas do you ever have a hard time deciding which ones to prototype nope we know we don't really have a lot of ideas we we just throw stuff at the wall and then it's so we just are we kind of operate on action as soon as somebody says what if we did this and we go okay and then there's a prototype within six hours and then that just sort of becomes whatever the next game is and so I know a lot of people have binders and binders full of ideas laying around their studios and stuff we are our Studios very Spartan there's just nothing we

don't know we don't have any ideas at all so yeah anybody else okay thank you very much guys [Applause]