Design in Depth at Supercell: Why Great Ideas Aren't Enough

[Music] welcome everyone my name is taka Collier and I'm a game designer and gamely that's what I sell I think it's not my fifth time here at GDC and I always find this super inspiring and it's now even more interesting to be here as a talker for the first time and why I like so much GDC I think it's always so cool to hear like people talking about games and how they approach those and and get like the insights and I kind of decided that I'm gonna do a talk where I go really deep into these details and give some inside there like like information like how how I approach and how we have approached at supercell today's design questions so first Who am I I studied actually Terry got physics as like my nature which is maybe a bit unconventional and background and I of

course like every terrible physicist tries to solve the theory of everything I also of course but I didn't unfortunately but I did some spend some time as a researcher postdoc researcher for me like games and designing game has been I call this a big hobby so board games especially I've done a lot of toast as a game designer and that's maybe Eclipse is my like most popular of board games and that kind of led me to mobile gaming as well and into supercell and I joined that about six years ago a bit more than that and happy dursa I said as game design and game lead so let's start about thinking what is game development and everybody knows it starts with a great idea your ultimate game concept and then it's all about sub just executing it right well at least in

my experience it's not quite so and I've started doing like in practice I think like game development is more like being in a channel you have your team and you have maybe some rough idea what's the game gonna be about what kind of game we are trying to make and the game is somewhere hidden in the jungle and you try to try to find it and if you think what what's the role for the game designer I think game is honor can be like a scout in the jungle try to try to find the best path and not get about getting lost and there are a lot of traps and you of course want to try to avoid toys as those as well so you could say that the goals for the game design is kind of find the best path through the jungle and minimize the time spent cluster and sometimes it is hard to know like when to push forward and when to change paths completely or or even leave the channel altogether today I'm going to talk about

my adventures in the jungle and I have here three different examples and these jungles they turn to be heyday boom beach and brah stars and when we started all of this project we had like a very small team three to six people depending on the game and yeah you're only in the beginning you have a kind of a rough idea what you want to do and but this is where we kind of ended up with the team in the first two in heyday and boom beach I was a game designer and in the broad stars I was a game lead for the nine first month of development but I tensed and left to help another team and I'm now currently I'm working on a completely new game team as a game leader but I try to pick like these practical examples from these adventures and go through them and maybe maybe you learned okay get some learnings out of Tom's toes yeah let's start with heyday

so heyday was released roughly six years ago it is a farming game where you build and manage your farm it is game where you produce goods and then you sell them and you have this order board which which you can where you get these orders and you complete the orders by tapping the button on the like the corner there and then you get a reward which is coin and XP and you use those resources then to mill your farm even even more nicer so when we when we started hay day our thinking was that okay there are like a lot of there are some farming games on mobile but at every more like ports from from other platforms and we wanted to make this native mobile experience and we also felt like the farming genre is actually quite cool but many games just don't fulfill that potential and yeah

regarding hay day just like three topics I want to cover first one he's like how we approach the UI design then about the balancing of economy and then trading and I'm gonna start with the UI part so I said we wanted to make this native mobile experience designed for for touch devices and when we started that we we had like a few core pillars in our mind for the UI so first one was that we wanted the player to be the main character and so whatever it means is that like the players action start from the player's hand if such as like planting seeds or harvesting crops and the we didn't want to have any avatar in the game which was quite normal at that point at least we wanted the game to be as intuitive as possible so follow real-world examples and for example if we didn't want that players would

collect coins from from fields for example and then when you craft goods the recipes try to follow real-world examples and also like us we we wanted to represent things inside the game and about these floating icons because we kind of felt that they it breaks the immersion so how we tried to represent the visual States in the game world I have few examples here so here's a chicken and here's a big and you can see the different states being from going from hungry fed and then ready to be collected okay I said we wanted to follow example but in case of week we had like a problem like how do you get taken out of it so so we wanted to be nice people and we're from Finland so the our ferocity of your solution was of course Sona so we have a pics online game which helps you to extract bacon make sense right okay about the UI thing so we have

this idea making a game for touch the touch devices but um in the beginning like how exactly it would work so I actually found this program a PowerPoint prototype I did back then and from there you can see like how we approach this so player taps the field and then you use at this tool and collect the carrots and that's it so it's it doesn't need to be fancy at least well this was the only only good tool we need to use at that time okay so how we tied it up in the game so here you can see play you're tapping to peel taking it a sickle and dragging dragging testicle to harvest it weeds and and the action starts from the player's hand as we want it so what are

pros and cons for dragging based you guys it is a very fun and natural way to interact with the world and this this bonus thing which is kind of side thing but I wanted to mention it anyways so when you are dragging it's possible to actually show some extra information so what I mean is you can see from the planting now player is planting the beet and when you drag the weed you can see the number moving and that's kind of shows you how many how much you have weight left can be convenient it is kind of hard to make everything consistent with dragging you still need some tapping and that's just reality there are menus there are buttons to confirm and also creating this kind of hierarchy is a bit difficult because basically everything needs to be in one layer and you can't become a messy and heavy for the UI okay next topic was about

balancing the economy for heyday and so what I mean by economy I mean the coins XP and unlocks so when in high demand player progresses against XP which unlocks you new buildings and then you use coins to buy those buildings in your farm and why this is yeah you should I should add that like player can play the game for several years and it's a long long journey so why this is important in this type of games player has these goals that I want to build the next building I'm going to do that and if the balance goes off the kind of game it loses its meaning so we felt that it's really important to keep the economy in balance so that every time you unlock a new building it is something that you can actually build or at least it's a fact you don't have too much coins or two little girls at that point and in

hey how do you get these coins so you have this order board which I mentioned and here you can see the goods you need from the order and here is the reward for that order so when you have any fulfill that you get the reward and now the question is like the starting point of all of all of the economy is that like how do you value the goods what is how much you get when you sell a karat how much you get when you sell a cream or for example so let's take a cookie as I example you get points on XP by selling it and how much of course you can just input some numbers there randomly and hope for the best but it might say make sense to have a bit more systematic approach to all of this so how we how we approach this was that we took the base fundamental resource to be time which is quite usually used thing so we said time is diamonds and what are diamonds they are the premium currency

in heyday but in here they mean more like a abstract quantity that helps us to balance and and keep everything in control but it is actually the diamond value for their stuff as well okay so time is diamonds but the relation in a day is not exactly linear propose they are not linearly proportional but there's this like a bit more complicated function that map's time to diamonds and I tried to draw their draw it there too you see and and what it means that on x-axis you have time and on y-axis it Maps the time to diamonds so let's take an example if something takes 100 minutes that would map to 7 diamonds if something would take 40 minutes that value would be if five diamonds okay now we have this relation between time and diamonds and how we used in heydays that we assigned for each good a diamond value and that's

based on production time it takes to make the good and also the value of ingredients that are needed to make the good I'll break it down so explain be careful is next so let's take a cookie cookie is made out of two beats 2x1 brown sugar and sixty minutes of production time we takes two minutes to produce so that put them up to one Diamond egg needs chicken feed and 20 minutes of production time and all the other toast would map to four diamonds brown sugar similary sugar cane plus 20 minutes of production time five diamonds and then you make the you bake the cookie in bakery which is sixty minutes and that's six times so when you put all of these together you have a value diamond value for the cookie and that's the premium currency value you would use

but also tells you like how valuable this is so how do you get from here to tech coin and XP value very simple we used just multiplier for the diamond value let's call it constant a and then the XP cost same way we have another multiplier B which you just multiply the diamond value and the XP value so if it's which certain values we get something like this for the cookie okay what is beneficial in the first place it means that all goods are valued in a consistent way if someone creates a new good and put it in their spreadsheet you immediately have a ball the values there and they let the player experience is consistent and then the balance is very easy to tweak because now we have this huge spreadsheet with all these goods and we can just tweak a and B parameters basically and and change the economy easily the diamond value has also used a

tool to generate the orders in the first place so well how we do it here's an example of order and it's consists of certain amount of goods and that we how we do it is but it's a that target diamond value for the order it could be like 100 we wanna order which is 100 diamonds basically then we generate this order by adding 1 to 6 goods so different good types and start feeling the order one by one with these goods until we reach the target diamond value and the target diamond value scales as player progresses so what kind of you get bigger and more more variety orders in the later in the game and then we spice up the reward a little bit so we shift the balance between coins and XP it's got a small addition to give it keeps the overall balance for the order

in place but makes a little bit more of these interesting choices for the player to choose like do I wanna make this order which has more experience has more coins some after does about this I think overall this kind of concept of abstract value can be a great tool to generate orders and also balanced economy and for player it creates this consistent experience and note however that real value doesn't for the player does it always to the abstract value so for example we have a products like brown sugar we do say you can it's used in many recipes so it becomes almost like this bottleneck resource in the game but it doesn't matter it still gives what kind of that those things can keep like a flavor into the game and in practice there's also always these complications and messiness

exceptions and they are unavoidable and I don't think you should become afraid of those but they are on top of the system you kind of have built so for example at the cooky actually it's not anyone diamonds in the game is nineteen but like I'd want to go into those details too much okay final topic in hey today I'm going to talk about this the trading and but we wanted to have a trading at the game from the beginning but we want it to be simple to use work with the core gameplay and then be a social experience as well so it meant that we didn't want to have a let's say stock market which is kind of faceless and maybe a bit like hard to use as well so how we approach this we each player has their forum and in the forum they have the a roadside shop and here you can like put stuff on sale and set the price and if someone visits your place they can just start just buy it from

your roadside shop and pay the cost so yeah it mean in truth through the friend list or whatever you can visit your friends and and check what they have on sale and then when you place goods on sale you selected of course the type you want to sell then the amount and then set a price price for the good and I would say like looking back I think said allowing players to set the price was a kind of a mistake there's two things which it leads to one is that players can like generate huge amount of money through the trading and obtain the other and yes and the players usually just put the max price for everything anyways so it kind of is a redundant but also tests like this interesting thing that we didn't think about at all was that actually wot's can use this system so you have like a what forms that can trade coins between accounts and create coins and all kind of hassle related to those there's also this button on the bottom

part of the UI which is where you can place advertisement for the order you have for the sale and what it means is that we have this newspaper in the game and whenever you like activate that tab then product the product or the sale will go to this newspaper and now everybody who's who goes to check the newspaper can see these advertisements and now this is a way how players can access not only their friends moves but like also people they don't know and get like a bigger feeling of the desam more trading happening and also buy stuff from players they don't know so now if you were tap any of those it would lead to the roadside shop of that of that player and then you can do we do the purchase there okay this is more like a general remark for the markets so it's kind of obvious but it the markets are

two-sided always and so you have the amount of purchases on there need to you have the amount of purchases and you have the amount of sales and they need to match its kind of obvious of course but why I am Hana say is is that it's actually quite hard to create this balancing balanced situation easily these markets are colorful unstable so do you have like an over supply or under supply so for example in case of hay day players want to buy stuff but it is kind of much harder to make them sell stuff and you need these two things to match how we approach that one in hay day was that there's like few things one is that we actually limit this so yeah naturally we limit the storage capacity but the additional thing is that we don't allow players to trash goods so only way basically to get rid of goods is to put them on sale so which will kind of drive them there and also we have a system which creates

certain materials as you metric a symmetrical e to the players and I'll explain what I mean by that so in heyday you have these upgrade materials which are very valuable for the players there are something that you get when you harvest or do normal actions and they they come as a random drops random drops from from those actions and how do you use those items you upgrade your barn and silo with with the items an important thing is that you need a set of items so equal amount of of those goods or materials so we divide the player to these groups we have four in here we are a group 1 group 2 group 3 they all get the same materials and they have the overall same drop rate but the percentages for these individual items they are kind of permutated which will make that some people will get more of

something and the others and will go faster it would kind of make make people trade in practice we have actually 27 of these groups because there's more items and there's more permutations you can go but you get the idea ok play yourself the idea was two players would sell the surplus and buy the missing materials I think this only kind of work to some extent so players are doing this with their friends but these items are quite valuable so in practice it's not always happening but at least some what's happening ok putting this together like what are the pros and cons worth of trading we did I think it's a it is a kind of easy and a way to sell and buy stuff it is a personal experience so you are always buying the items from someone else which is important you can also discover other players through the newspaper and at the same time you can visit their farms and and see what they

are doing and maybe get ideas for yourself as well in practice the newspaper is not always the best place to like look for if you are looking for something special it might be hard to find it and also for active traders the coin balance they can kind of ruin it by just doing a lot of trading and then we have to bought what problem as well so this is created a system where boss can actually exploit the system okay I'm gonna switch now to the next game which is boom boom boom sorry it's it was launched roughly four years ago and boom is a kind of building battle game where where you have your own Island and you have your own base in the Thailand and and you have a map where you can find other players and bases and attack those and if the attacking you have a you deploy

troops on the shore and the troops have an AI and table move but in addition you have this gunboat which you can use to place news players to guide the troops and use different of the gunboat abilities to help in your attack so there's like a bit more control perhaps than in some other games with with the attacking so when we started boom beach they've kind of one of those initial ideas is was that when we wanted to do this was we there was a clash of clans of course and we felt like there's some more directions we can still explore within this genre and one of those like main things was that clash of clans has this matchmaking system where you just press a button and you're matched against opponents or an or then you have the single-player map and it's very convenient and easy to use but on the other hand you kind of lose this feeling of a bigger world and I know they are also separated and the other thing was the idea of the controls that we we wanted to give more controls for the

player in the attack with a flare and other combat abilities so I will say those were kind of the main reasons we wanted to explore this channel and okay I'm gonna talk about regarding boom Beach about the game world and the map how we built that and then about how we balance the combat so let's start with the game real demamp part with this one I said we want to create this more immersive game world where player would feel it's not just like a UI but something else we wanted to introduce exploration element and mystery with that so there's always something hidden in the clouds and and you can find different interesting stuff it creates this nice feeling of progress when you when you play the game we also wanted to combine NPC and PvP enemies to a single system which is your map but we wanted to avoid these typical problems that lie in these like a PvP map based

games and I'm gonna say a few words about that so here's a Travian it's a super super cool game with a like with a map where players are actually placed and they have like a physical coordinates and and if you attack like a depending how low how far they are from your home like it takes more time so these maps can be like really nice but they have some problems like obvious one is like when new players join scenes where you place them and you typically T's like games have a one server which can host say 20,000 10,000 or 20,000 players and there and then you split them to these instances they are like said they can't be scaled so well not in space but and also not in time so when the time moves forward comes the dynamics are some many times quite unstable so power power players dominate easily over time and if the world can

become like a really hostile place so we wanted to avoid this but still create some kind of map or at least feeling of map and kind of the boom message is really a kind of a fake map but it does its trick to some extent at least so here's the map design which the map is divided to these very oddly colored regions sorry about that but idea is that these correspond to these clouds that you can unlock and you up be under those you can find these different kind of nodes what I be my nodes are like a PvP opponents like treasures but and here's a map all with all of the elements this is not like a real in-game picture but like this you can see all the different elements or it is from the debug but yeah it's so there's a resource basis PvP bases different different type of things and what I how I what it should be kind of think or see

here is that it creates this element of scouting and planning because you can see your opponents you can tap them and sculpt them and chill see how their base is built and then decide how I want to approach this base so it's more like a puzzle almost you can attack it multiple times choose your army composition try different things so create sex coating and planning element to the game okay how do of course with this kind of map thing the question is how do you find these opponents in the first place and Tessa open PvP opponent between match makes on your map based on the mid middles which is like the trophies in some of the games and taste like to you ways one is through exploration so when you uncover the clouds and there's a pvp node we must make opponent there and if you clear it then the node will be empty

and we can then create new enemies from true invasions so what are these invasions they are like a way how its spawn enemies on your map or much make them on your map okay this is like a clear challenge with this one which is that with the invasion system like what's the right tempo for these invasions some if let's say you are really active player and you play the game a lot you clear out the map super fast and then you don't have anything to play basically so that's why but on the other hand if you are a player who's more casual and you play it every moment then if people just like pull the map all the time with with opponents it might be like too overwhelming so to talk a lot of it was super hard problem to make the invasions work such way that it kind of pleases everyone and it doesn't at that's that is a thing that we can't get the work to for everyone optimally and that we have to make some compromises there but we still want to keep the innovation algorithm same for everyone

because it is a competitive game that it would be unfair to change it for different players but it is a dynamic in a way that if your map is very empty we spawn faster and if your map is full of opponents then it's kind of start slowing down and then there's other other problem which is that how you actually get rid of these opponents you can beat and this exclamation mark on top of the base and this means that the base is something that you can rematch make it praise the immersion a little bit product was something we needed to include there to make it work okay how we type this map to the story out of the world so it's a kind of unusual setting where each player has their own map and they are at the center there in the center of the map and everyone else are actually like enemies so all the other PvP opponents are represented as a part of this evil blackguard movement and you are fighting

against those so you get to be the good guy and then these empty nodes we represented them as like a native villages you try to protect so when the invasions comes they take over these Willet villagers and and you try to protect them so it may be weird but an unusual setting okay next second topic with Palm Beach and it is about combat balancing and I wanna motivate this a little bit so in clash of clans you have a lot of walls and think but in quite quickly realize that we don't actually want to have walls and boom beach there are several reasons and one was that they kind of how we how the attacking works with the players and everything they just didn't and with one direction of attack they just didn't work as well in the combat and with that because they are like a huge amount of content in clash of clans and provides this long preparation for yes we needed to have

more content in somewhere else so we need to create like a lot of grade levels for the buildings and for the troops so this roughly 20 upgrade levels for most of the troops and buildings but now the problem is like we have such a bit big journey like how to keep it all balanced and so I gonna explain now the some of the things we did there so first of all how to boom beach progress works is that you have your headquarters you operate that and it will unlock your new troops you can both abilities new differences and new landing crafts so the landing crafts will fit your troops and if you upgrade them and get more you will get like bigger army that you can use in attacking and the defense's can be of course upgraded as well and they are tied to the headquarter level now you have an armory which is a place where you can upgrade in troops and gunvolt abilities and they are kind all

interlock to the headquarters in a way that it limits all those AK max upgrades for each of these ok I'm not going to take an example of riflemen and cannon and quickly show like how how these work so we have a little one and you have some amount of hit points and damage for those troops and level two it increases and it goes up to level 20 nothing too special here note that they are kind of these increases are always a percentage so you can you know that each level is like a certain amount person sees more than the previous level and they stay the same through that journey and in similar way you have the defense's they have some hit points and damage and they go up to the level 20 and here we have also some percentages they can be different and they are different for different buildings and troops actually just to make the gameplay like work as well as possible

okay no we want to keep like have some control of the balance we would want to have a map each of these like levels and different troops to certain value and for the attacking troops we call it offensive value and for the defense through defensive buildings record attack a defense value so that's the offense value and for the rifleman the starting it's kind of this starting point for everything something where you you can compare everything against those so level one rifleman okay we just said it it's then it's the target value for that now you can take the cannon and see like how how they compare so you can say the cannon has certain defense value and now you put the rifle man's and all the units play against the cannon and just by play testing basically you can see but okay the cannon is six times more worth than the rifleman so you could put place there like a value of 60 and then you can play the game for 40 max levels as well and find out that okay the max

value is like offense value roughly 47 and you know because they go like percents twice for each level you pan can't be sure that all the intermediate levels are you can kind of extrapolate for these two numbers and same thing for the max a cannon level you extrapolate and get like it like depends on offense value for all of through the progress okay how we use this so there's like a few things how we can now use these abstract values first of all you can try to map the opens defense balance through the unlocks so this is like a craft where you can on the x axis x axis you can see headquarter level and y axis you can see this abstract total power in your defenses and in your offenses and the blue lighter blue is the orphans and the darker is the defense this takes into account all the like the max upgrade levels at certain so it comes

basically more you'd spreadsheet which takes into account all the upgrade levels that you can do in any level and all the unlocks and if you see like okay defense value is keeping too much compared to the offense value at certain level we could perhaps unlock a new building there so it keeps helps keeping everything in balance in that regard and also note that the offense value it starts like a bit higher on their conduct early stages but dips down later and thus takes into account kind of the skill of the player which when when you progress to the game you learn to play the game better so we have to kind of compensate it a little bit in there there is this extra challenge with the map we did which is that we wanted to have the PvP and NPC integrated in the same system and it's easy to get enough content from the PvP side you'd you just have to make people's but then on the NPC side it's harder the problem is like how to make enough of that content which would be in balance for the players not too

difficult not too easy and don't creating thousands of levels by hand and balancing them was not an option and by the way till they don't need to be perfect balanced but like having some rough scaling for the NPC levels would be good so how the defense value helps here is that firstly you can create any level from level editor and then you kind of get the difficulty automatically so with level editor you get this you can do something like this and now if the question is like okay how hard this is but you can just check the buildings one by one some the defense values from these buildings and then you get the total defense value for the base which would be here 3562 which is it's not the real case I just made up the number but you get the idea and this number now we can use to tell like how keep some rough

idea what's the base total difficulty and we can then match it to the player progress and of course note that the layout affects the difficulty in practice but uh but it kills if you keep like a rough idea good enough in a way and then you have a how we used is also and this actually is quite a big part of the game at the moment and we use this defense follows a tool to generate levels so we began existing levels that sank someone made by hand we set a target defense value which usually is higher than their normal one or the initial one and then we have this algorithm that places you building new defenses maybe remove something upgrades them modifies the layout and we can do those tricks up to the point where the P get get to the target defense value and then voila we have a alternate version of an existing level and this defense party was also

used to I'm not going too deep in here but like that we have these events which are also like fully generated by beyal just algorithms and they're the same thing goes but like now now they're not many existing like a starting level here it's just like piston piston bytecode but they're still like a target defense value and we can control the difficulty for this event to some extent of course okay my last game I'm going to talk about is pro stars and I it is a game which is in beta still and it is a arena action game kind of like a bit like a mobile something where you have a team versus team battle three heroes on one side and in game you can play different brawlers or heroes we call them brawlers when starting the

project like one of the guiding ideas was that we really believe the taste of people want to play this type of games from mobile but they were kind of the perfect game we felt was missing still and we also felt like we share sometimes should be really really short and these were kind of some of the ideas why we why we started the project today here I'm not going to talk about two things one is about the hero progression and then about how we tried to develop the controls okay about what I mean by here appropriation I mean that there we wanted to guarantee lair has like incentive to play all the heroes so brawler's basically this is like a typical problem with these competitive games and and I call it favor through a problem it's really bad bird but couldn't come up with anything better basically what I mean here is that you have a scoring which which

punishes for losing obviously because a combative game and it's usually more effective to just to learn play one hero well or one deck whatever is the game and it's also better to concentrate upgrades on this one single hero if there are any and so you end up sticking with your favorite one what is the problem of course it's not it's not in a very big problem if game can be really fun as well but theoretically if you bring out new content it can be it can feel irrelevant for the new players and also updates as well and also the other side is that actually player might want to do something else they can get bored in these type of games because they can't locked into this they favorite favorite thing but it's just too risky the punishment is still too big so this happens in many games like I guess in hearthstone more many more buzz clash royale you faced a similar issue a little bit how we try to address this in

brothels was that we had a Egypt for each of these heroes each each of these brawlers they have their own trophy score which is used in the matchmaking so you can see like here's the trophy score for four drop here then we just like we we still want to keep the overall progression for your yourself and that's just like a sum of all the prologue trophies and which is shown here in addition there are different game modes and levels which favor different type of rollers so it also pushes you to try and try to apply different driven progress in different events an upgrade system makes it hard to concentrate on just one character there's one like interesting detail about the trophy progress which is that it's kind of personal and what I mean by that

oops sorry what I mean by that is that your trophy changed only depends on your own trophy score so test this table which tells you like if you're you are playing with certain hero and it's a certain amount of trophies you will get a different amount of trophies if you win and lose depending on what's your personal progress and the opponent's progress doesn't matter here at all so it's not like a zero-sum game in in many game zero-sum thing like in ello which is used in many games if you have okay after those about the system I think the positive thing is that it come it's easy to learn to play a new hero without any stress and it creates this fun and fast progress after unlocking a new hero there's no like punishment to play play new new hero that you are not or game it

does have some complexity to the UI which which can be seen as like a downside I guess okay final topic about process is the how we how we approach the controls and as you see there's a tiger in the channel and so we kind of thought that okay yeah we can do it's piece of cake well not if not piece of cake but at least something something easy no I'm not easy but doable but it is a challenging problem to do a like a our action game on a touch device and I'm selling things like there's no perfect solution for this one actually so the starting point for us was that we wanted to make an action game which is in landscape and the attacking direction from would be from left to right and how do you control would be by tap to move so that's kind of the easiest way to rate

the control and then the skills would be triggered automatically and so by some sudden triggers basically when we quickly rejected this because we worried that there's not enough depth in the game our second try we kept the same landscape mode and an attacking direction and we had the tap to move so we had a now out of fire which I mean that when enemies come close enough they in the range the character will start shooting automatically but then we had a skills that can be activated by by just tapping buttons and then you would have a cool-down ster we rejected this one as well and the reason was that there's not like feeling of of having not enough control for the characters third

iteration was a bit similar tap to move all the fire and then you had a skills that can be activated and targeted so now the targeting is anything but uh again still not enough control yet lots of stuff happening okay at this point we'd figure out like okay let's let's think about it that there needs to be more control for in the shooting part especially one that doesn't need to be but it just feels natural in this action that game that you you can control the shooting and we still get there at the direction and started prototyping and we tried all kind of things like a different joystick options shoot button with our auto targeting and many others in the end after prototyping some nth iteration can't count anymore we actually end up with something quite

different which was you would have a drag to aim so when you drag the finger you open up this like a aiming cone and yeah yeah and now the our direction was like a portrait mode which was a big thing so we figure out that okay let's let's let's change the portrait because it's a kind of net more natural way to hold your phone and the attacking direction became from bottom to top so time to move and yeah that felt pretty good we were like yeah let's ship it this is it but we we proceeded to company playable and just make sure we added three different versions to compare and some of them so few of those Hadlock are so different versions of this and also like a different joystick options and we got a feedback that like some really preferred the joysticks and some prepared like the control Metro we had

first thought so it was time to time to go the beta and how to go there so we these are can go there with two options first use the drag to shoot tap to move and then second have this joystick to move tap to shoot so these two kind of from those playable some some versions of those and then we divided the group of players to do two amv group and just see like because we honestly couldn't say like what's the best best veggie plate and yeah what were the results from this a be test very interestingly there was pretty much no clear impact on metrics and I think like this is something that's happened over and over again like we actually do very rare like really a be testing in the company and especially for the gameplay things but when we have

done it it's like you said it's like not clear right but so okay and you would ask him that something like fundamentalist days would put show on the matrix but there was we had like a possibility for the players to switch between these two control methods and in the game because we felt it's kind of fair they're fair to give them the possibility and interestingly people switched more to the joystick over time and we don't still know exactly the results but you can speculate that maybe it's related that it it might be a bit more effective in some competitive play and maybe it's also more familiar for many players to use this kind of just joystick that type control middle so the decision for them was to go all-in to choice tick and what because of that it made sense to change the mode to landscape again but the attacking direction stayed from bottom to up and this is how it looks now so it is a kind

of a glow stick it's very recent update and this pin like already and I've had some iterations on top of that but basically this is it so there's a duo stick with this has this kinda like a little bit innovation on joystick side we call it trip stick it's hard to explain here but then we have a choice T for Amy and in here if you drag you can visualize the aiming and if you release then it shoots we have also option for tap to shoot and attempt to alter shoot which is what pastor ray to I should be the auto aim and then there's the joystick for for the super which you released in same way you release to shoot and then you you can have possibility to do the tap to out the shoot okay so what's next for bro who knows but we'll keep on exploring

okay time to wrap it up so some conclusion time so I kind of made the case that gave a double band can seen can be seen as I exploring a channel and antenna interior in here game designing is about scouting and then I talk about my adventures in here in the channel know if all of these examples have been quite different but I think like there are some some general themes between those and one is that you can use this kind of abstract value as a tool and if excused in Haiti and in Palm Beach the balanced economy and balance the combat and it can be like a mapping something to abstract value can be used in many many ways and I think it's good to remember that every jungle is different and the game developer is full of surprises and especially if you are like innovating and doing new things where

you are lacking clear references and something that can first look like a really easy can be actually very difficult in the end and good example is pro stars for example and control stare and perfect solution on paper might have unexpected consequences in the end as indie here witnesses when he picks up the statue okay thank you okay I think we have some time for questions if you have any either it seems like rule stars has been the project that you've had in visa for the longest as a designer what does it felt like to make those major a serrations whilst the projects actually live so uh I've been actually like I've seen the

early development like a part of that but and then I've been helping other projects so I'd be no lack of following the side and helping from outside a little bit but I think it's just natural like you try to make the game work as well as you can and I don't think bi is anything like a you know very different you it's just you develop it inside the company somebody and put it out and then you keep on developing and try to make it as best as you can so I think it's natural process of up game movement right see time field is so that you don't feel more pressure on the design decisions oh yeah of course there is a bit more pressure because tech like a community and I think community has to be taken into account and it's really valuable also to kind of get their insights but I could kind of think that you also you get so much valuable information there that it helps but if there is definitely more pressure yes thank you them hi thanks for the jungle metaphor of exploring I was and also for the detailed examples I was wondering when you start out new projects do can you start off from this small explorations of a single mechanic

or do you need a high-level concept first what do you do in super solo ah yeah I think it depends but over I think it's good to have some overall idea like what you're trying to do like maybe what's the who would play the game and what type of experience you're building but uh there I think it's a combination sometimes you can kinda have some like a really good idea that you feel that you can build a game on the on top of that but uh I think still I would say like a big big overarching feeling like what type of genre you are doing and what is maybe the audience for the game I think that's still the most important thing thank you yeah alright I want to ask how did you if you can go into more detail about the decision of changing the like latest changes because as you know brown stars is now the landscape mode and it needs to be a per tournament for long if how does this was made or what would you guess would be a good time instrument for like this is the enough time we will try with this try for portrait and now we will go to an skate because we have to change it or if you go in and kpi's thank you yeah okay so so the question was that like

what what was the kind of crowd decision processed okay yeah what's the decision process and when it's like basically enough will changing this and will move the other I guess like it's also always be present at how the team feels in a way like they they are doing the decision and yeah there are many factors to it I don't I don't think you can say this like I you have to very certain I'm time enough but I think here like the big thing was that realizing that okay joystick is seems to be like how many are playing and then it can't make sense to go landscape with landscape because joystick is more natural in that position and then it kind of escalates to many other changes and voila yeah it's kind of organic process basically thank you hey thanks for your talk a lot of cool stuff you mentioned that there was no clear impact on metrics when you a be tested top controls and joystick controls but with and with this new version you change the landscape and dual stick controls but you removed top

controls right yeah and was there some impact on matrix because you like nucleus they are not I don't actually even know to be honest so I probably something but I don't know even like which direction yes oh yeah thanks hi my question is is there a specific reason that you made the conversion between time and diamond not linear what sign go there okay that's good question I think like yeah there there is and mostly like to test two things to that I think like you can think so in saturated if it would be linear like the like very long production times would be way too expensive but I think the other better way to think about is that like you can think there's two types of times for the player one is that you are spending when you are actively playing and that's kind of those short session timers and that's more valuable for the players dantos like okay I put the game down and go to sleep and it eight hours so it kind of makes sense to value those like

shorter sessions more that's how I think open it Thanks yeah hi thanks for the told thanks for all those detailed informations I would like to go back to heyday from the beginning of the talk you mentioned that upgrade of warehouses all those currencies were developed mainly to enhance trading yes as far as I remember before getting this idea how how did you approach warehouse upgrading it was just like the normal normal material with all those experience if you will do it the same as you did so would I do it same way would we do it the same way now I don't know I think what it came all came at the same time pretty much I remember if I remember correctly so I don't think so basically the trading and yeah I can't remember the order I think it was pretty much like the same things like locked into a place in that at the

same time so I don't know if what was the actual order of up things happening there yeah I don't know if we would do any changes I guess like sometimes it can feel a bit too frustrating to I got credit so if there would be some like I don't know that might be something to think about like yeah because I played Haiti and it felt quite frustrating to me from time to time because I didn't I didn't have that many friends playing at the same time you know it was kind of hard yeah I think like it sometimes it's too hard to do the trading photos are pretty materials and the idea to how to facilitate the better I think it works in neighborhoods and works with friends but if you don't have that access point and it is a bit hard at though I agree yeah okay thanks a lot thank you Hey Hey I'm thank you very much for your talk it was the start playing roll stars and they're like very beginning and I always was wondering that a these guys they're using like two different control systems so I was thinking that and which was a

controlled share I mean which option was much more popular in in players and also was wondering if do you know like if players that was using one system were better than the other or the higher winner rate or if you check this data oh yeah I drove it well in the beginning of course everything start the population plays 50/50 because we did like the AP test in splitting it into two parts but then it evolved so that more players like adapted the joystick option and especially in like a very high level players and I if I understand correctly and I it is yeah it might be R it's a bit more effective because you can talk depth so quickly to shoot but yeah that's what that's for how its kind of fun but I don't actually know even the numbers to be honest exactly so I can't say okay thank you right here hi for the MPC combat generation in boom beach yep

you mentioned that there is a method for creating a template and moving values until it matched a certain defense value that you wanted is that how you end up doing the entire game is that all the NPC are based on the templates or there was some dynamic like just freeform generation that was non template yeah basically like this just like a premade level if there's some like several hundreds of those actually but then those are used as like a templates to some player might face like several thousands and PC levels so those are used as the seeds so first time you meet them they are like just a basic level but maybe next time you meet it it's kind of harder version of some of those earlier levels with some modifications so we kind of use them as a starting points for the generation okay all right my questions going back to heyday I like how you are basic my map diamonds to time but then how do you determine by how much time certain things should be you know like corn should be axe you

know other things should be why how do you know that old this is costing this is costing too much of Claire's time first yeah it's a good question I think how we approach it I don't think there's any like a right answer but that the time is kind of the player experience and we try to make it like how many how much things take time it's kind of to optimize the player experience and then they are kind of mapped to these values automatically so we we set the times by justice or that it feels good and but I would say like there's a bit of this trade-off that you want to make some things like a really fast in the progress like we eat for example super fast to produce but it kind of cool causes problem later song a little bit because it's so you can so fastly grind it and but it needs to be fast because it's the first first crop that you unlock thank you thank you I also question about hay day let's say that you figure out that the ingredients for a mission add up to 21 Diamonds do you then give the rewards exactly for 21

diamonds or do you vary it a little bit so that because it feels like if you always give exactly the reward that are balanced then it's almost doesn't matter for the player which one they pick so the baseline is is always the same but we spice it up a little bit so we have this thing that the XP and coins they are kind of the balance is shifting so sometimes you can get like more XP from the same order but then you are getting less of coins and then add other way around so there is this kind of randomization on top of that which shift is the balance between but like the combined value stays the same and ads come flavor to the game in a way I agree it could be otherwise too dull and that's why I kind of made it yes goodness okay I think we are out of time so thank you