Balancing the Economy for Albion Online

[Music] yeah hi my name's Matt I've been a game designer for like ten years now I did eight years in Reykjavik Iceland with CCP games on EVE Online which you may have heard of and I spent the last two years in Berlin Germany with the company called sandbox Interactive making Albion online which you probably haven't but that's what I'm gonna talk about today so things I'm going to talk about today I'm gonna try and explain what this Albion thing is that I make at least to the degree where you can understand the rest of the talk I'm gonna explain what I mean by economy balance and why it matters and then I'm gonna kind of walk through our balance process or like kind of economy set up process for Albion I talked about setting goals why they important how we setup the relationships

between things a little bit on doing the math filling everything in and then kind of what happens with me ship how we deal with feed about how we kind of iterate on it as a result I'm gonna try not to talk to you fast I've been a little bit jet-lagged so bear with me thank you to explain Albion I want to briefly thrash through the family tree of Western mmo's ignoring Asia because it's convenient not because it doesn't matter I just I don't need it for this but we love you Asia so we start off with multi-user dungeons back in the kind of late 70s early 80s basically a way of allowing people to play versions of Dungeons & Dragons on university mainframes with their friends and we kind of build from there in the late 90s we get the to kind of sort of beginnings of modern MMOs coming through we get EverQuest and we get Ultima Online EverQuest we can already divide this tree branching into now

EverQuest is kind of what we refer to as a theme park game or an early version of a theme park game which is largely about gaining experience and leveling I've been getting lutein progressing through game-2 the endgame ultimate by contrast is what we refer to as a sandbox game it's much more about kind of giving players a world and some tools and letting them kind of interact with each other and figure out themselves how to have fun to a large degree so then they run for a while they do okay and stuff early two-thousands we get enough of next iteration yeah world of warcraft really kind of consolidates the the theme-park approach it adds an extensive questing system which really gives you those rides in the theme parks and you know you get on your quest and you go do the question is wonderful and then you're done with that quest and you move to the next quest very much again progression focused and getting to the raids kind of focused Eve by contrast is kind of Ultima in space it's still very focused on players interacting with each

other and kind of emerging gameplay and kind of seeing what happens in the world probably the biggest departure there is the Evo single sharded most mmo's do what we call sharding they split the world up into multiple copies of itself and players connect to a particular shard and that's where their character is and their friends are and they play on that that copy of the world Eve everyone connects to the same world space the same instance essentially at the world that's kind of quite a big step for sandbox game because it really means everyone is interacting with each other as much as possible from their world of warcraft leads to basically all the arrows in the universe well the woke up makes a lot of money as you may have noticed and everybody else wants some of that money so they go after that and that leads to a lot of our Mo's and on the sandbox side we kind of get dark for which you may have heard of which is a kind of a more PvP oriented version of Ultima and then not very much at that kind of scale and Albion fits in over here we're inspired

by evil inspired by dark former spied by the original ultimate that's kind of where we fit in so we are an MMO we're like World of Warcraft and then that we're a shared persistent fantasy world but we're not like World of Warcraft in most other respects hopefully that gives you a sense of where we go I was gonna do and a slide explaining what is this Albion thing but then I realize we've made a video mine is a sandbox MMORPG in which you get to write your own story instead of just following a laid out path explore a vast open world consisting of different unique biomes everything you do has an impact on the world as in our player driven economy every piece of equipment is crafted by players from resources gathered by players the equipment you wear defines what you are switching from night to magician is as easy as swapping your armor and weapon or be a combination of the two venture out into the world or into dungeons and face

aliens inhabitants and wildlife go on expeditions to encounter even more challenges fight other players in open world clashes fight over territory or city control in tactical group battles relax by retreating to your personal Island build a home grow crops raise animals join a guild everything is more fun when you work together leave your mark in the world in Albion everybody matters yeah so that's kind of basically what we do it's single Charlotte suppose sandbox see if a player oriented three kind of things that I'd pull out that I think make us stand out Albion is we regard ourselves as properly multi-platform we were leasing in July we hope on PC Mac

Linux and Android tablets with iOS following soon after the mobile clock we're targeting tablets and be targeting two gigabyte of memory cuz it's kind of intensive but it's not just like a companion app what you play on the tablet is the same client it's just a different build of unity connecting to the same game well with the same stuff you can go raiding and pvping on your iPad if you want that's led us for example to our art style you saw it's very kind of simple I think looks really good still makes us think a lot about accessibility and about complexity of the game kind of trying to bridge those two audiences the traditional Alamo crowd which is kind of our primary audience but also bringing it to the mobile arena as well another thing that we do we focus on small team PvP a lot of sandbox games particularly something like Eve you've probably heard of you know five thousand people fighting out in the system which is cool and we still have the opportunity for that and our open world PvP but in terms of conquering territory we we take a break

from that direction the way you take another Gil's territory is you you pick a one of their vulnerability slots in their time zone and then you bring five players and they bring five players and we wall it off all the territory off into like an arena and you fight it out with the equipment both sides of board so even though there is opportunity to go big if you want to be you can be a small guilt you've got five people and if I could fighters in a couple of you know logistics and crafting people you can claim your own corner of the world without needing to bring hundreds of players so that both kind of brings something a different direction to some of our PvP and also informs a lot of our thinking because we have to think about this you know tightly-bound small-scale engagements the third thing we like to wave around is this uru where thing which works very well for us obviously it gives players a lot of choice which is good for them it also means that kind of the more macro level we've got like 45 base weapons and 27 armor sets I think trees and armor items wish and

they all have spells attached and a choice of spells and you can you can mix and match those completely which leads to a lot of variety in PvP which is what we're going for it means that every fight can be different because players are constantly looking for that improve build and that counter build and all this kind of stuff so it keeps the PvP as fresh as it can rather than like there's no class this has no restrictions on this stuff so you're not just like oh it's a monk he's probably running one of these three bills we know how to deal with that you see someone like he's wearing a plate helmet and cloth hues what's going on here it just leads to a lot more variety and interest there so that's kind of that's the game we're talking about hopefully it gives you almost enough of an understanding to understand the rest of this talk I'm gonna get into the stuff get into the meat of it now what is it calmly balanced and why I have a philosophy background so I like why questions I also find the answers to them really useful and kind of informing my thinking hopefully it will help you too so I'm gonna be talking about yeah

what is it comedy balance why do we care about comedy balance I think these important questions so economy balance we take sandbox a pretty expansive view of economy basically all the macroscale balance in the game we don't we're not worried about stuff at this the economy stuff with like you know does the axe do more damage than the sword but like does the level 1 axe be more damage from the level 3 acts and how much that's already getting into economy territory because it's a it's a power scaling issue so yeah Khanna means all the macro balance it means obviously resources and money are silver and are various resources you can collect it also means the progression system we have this like tree structure when you earn Fame which is our version of XP in different areas of the tree like you can you know you use a double-handed sword and you get better with double handed swords but the balance of that ties in the balance between like I was saying equipment tears ties into the economy balance because it's all of a thing our farming system for example you know everything else of that kind of

scale of the game while we're trying to balance resources and and macro level stuff against each other all falls under the aegis of economy balance and it all gets approached generally the same way the same kind of process which is what I'm going to talk about in a bit why do we care I like this question I think this is important and then do some theory we want our going to be good right and I don't like using the word fun because I think it constrains too much what we can be doing with games like if you if you ask someone who's climbed Mount Everest if they thought it was fun they would probably say no that doesn't mean it wasn't good so we I'm gonna try not to use the word fun too much in this talk and like because I'd say I think it's too constraining for what we can offer at the game particular when he get into virtual worlds like we do and we think our economy balance can make a huge difference to whether or not our game is good so we do want to think about a case and we do want to get it right so let's talk about motivation I've been relying

for many years o use it a lot on the eve team and I've been using it as well as sandbox on a model that I borrowed from or we borrowed from Scott Rigby and the folks in immersive I think it's cause doing a talk later today so she goes into him cuz I'm gonna try and skim over it I'm gonna get some of it wrong I apologize that hopefully gives you enough to understand where we're going with this so in this model we have we have intrinsic motivators for starters you have autonomy feeling like you're in control of what's going on relatedness feeling like you're connected to other people and mastery feeling like you're improving in skills and these are our intrinsic motivators these are things that just make people feel good and if you can get them doing all three of these things at once then they're pretty much guaranteed to be enjoying your game the real tricky one for our Mo's and other long-running games we find it's usually mastery because mastery is not being good at something it's getting better mastery consumes that's why you have to spend so much on on content in theme park game for example because the content is consumed as you go once

you've beaten something you've solved it it stops being interesting from this point of view so yeah if we can hit all three of these then we're probably in a good place by contrast we have extrinsic motivators this is basically any kind of external reward so making numbers go up particularly numbers right into your character your you know your level your XP your money whatever and getting stuff everything to do with loot and that kind of thing and the he seems the human brain really likes this stuff it likes seeing numbers go up faster and it gets very motivated by almost getting what it wants so if it almost gets that amazing loot drop its after combined with the fact that human brain is awful at probability means it's very easy to trick it into thinking was almost there and to keep trying and this extrinsic stuff is very compulsive it's not necessarily enjoyable the people will do it anyway and they will keep doing it until they run out of things to improve or they hit

some kind of breakpoint and look at themselves and say why have I been grinding walls for 60 hours and they can walk away from your game and never come back at that point so it's kind of a fragile way of people getting people held in so yeah we have these two different things intrinsic and extrinsic and when the two compete when there were different activities with different motivators competing for players attention this kind of tends to happen the extrinsic motivator always wins and the other way of breaking it down is saying the intrinsic stuff is what your players will enjoy in the extrinsic stuff is what players will do and your economy obviously is all basically your entire extrinsic motivation system another way of defining economy for our purposes is all of the the extrinsic rewards in the game your economy is basically one giant incentive structure that tells your players what to do is they go looking for the best ways to level up whatever that best way is is essentially what

your game is telling your players to do so if we look at a traditional MMO we make a simple Venn diagram we break it down three areas on the left that is the lap for you yes we have the blue bit you have your intrinsic motivators that this is our age of the game the pure blue that are intrinsically motivating but are not extrinsic motivating this is wasted potential this is all the bits of your game that are good that one is gonna play because there's no reward for it in the middle we have the bit of your game that is both intrinsically and extrinsic rewarding this is the good bit of your game the bit of people will play and enjoy and on the right we have to grind this is all the stuff that's purely extrinsic this is killing a million wolves because the wolves are the most efficient way to level up there's no kind of it's not good you're not enjoying killing the millionth wolf particularly but it's very efficient generally speaking the problem here is missing mastery you've figured out how to beat wolves 200 hours ago you're not learning anything it's

just kill a wolf kill the wolf kill wolf so yeah this is the bit of the game that is usually called the grind this is why we're grinding it's because it's grinded because there's no intrinsic value in it so what we are trying to do as a connolly balancer is we're trying to do this we want to make it so that the intrinsically rewarding parts of our game and the extra interview rewarding parts of our game aren't as much as possible overlapped and in particular in a more subtle level we want to ensure that the most intrinsically rewarding parts of our game are also the most extrinsic awarding parts of our game and vice versa so that as players try to optimize that process they try to find what's the best way of doing this the game doing that takes them to the best parts of your game right it calmly balance success you can do that so yeah that's basically the theory let's move on to the next bit so we have that we have our wise let's talk about how we actually go about doing it what I'm gonna do here basically is try and take you through a

kind of a simplified and slightly idealized version of an earlier iteration of our economy balance because it was still at that point simple enough to explain in an hour hopefully like the important thing for me here is the process right it's how do we do this the fact that ax takes such two resources the why of that is probably not nearly as useful to all of you as how we get to that number and so that's one I try and focus on and we're going to start off with goals if I had to define design in a simple sound byte array I would say that design is art with goals design is art that's trying to do something in particular and the goals are kind of what drive that so let's talk about some goals well let's talk about why we do goals more to the point a few kind of bullet point reasons I find it brings a lot of clarity and a lot of honesty to the process it makes you pin down upfront if you start off with your goals what you are doing and why you are doing it helps them to

understand kind of where we want to go with this feature and some way that just writing anything down will often clarify the thoughts writing your goals down and getting them pinned down somewhere and can really kind of give you a good foundation to build off you know where you're going and what success is gonna look like when you get there I found it's great actually for team harmony the eve design team in particular is like 15 or so people and did have disagreements sometimes about how to do something like where we're going with the future how we wanted to do a particular thing and the more I was thinking about goals and applying that to my design work the more I realized that most of those arguments were happening because people had different goals and we hadn't stapled what the actual goal for the future was and therefore that talking past each other this guy talking trying to design of feature for veteran players this coach wanted a designer feature for new players and of course they can't agree on anything because they have different goals in mind so getting them pinned down and everyone on the same page can avoid a lot of those things and when you get into those discussions if you stop and say hang on what do you think we're

trying to do what do you think we're trying to do that's often enough to actually to diffuse the entire discussion or at least bring it down to a level where we're talking about the same thing to each other I've also found goals great as a supporting structure I kind of a simple level just having something to go back to when I'm when I'm stuck or I'm struggling I'm not sure where we're going next it's going to look at the goals where we're going with this feature how do we want to get there it's really helpful and kind of at a broader level goals bring you into a kind of your area of like constraint driven design you're narrowing down your possibility space by specifying what you want to achieve and that limits your options which for me makes it much easier to design something good because I'm much more focused on how exactly do do this rather than where and this great big blue sky are we gonna go with this feature so laying out those constraints early on and kind of pinning down the options I find really useful to taking it to a better place in principle obviously goals should be perfect and immutable and laid down at the beginning in stone

tablets in practice that doesn't happen we find that some of our goals are wrong or we want to do a feature in a particular way we really know we want to do this but it doesn't match our goals and it's fine I think to go back and change those goals so long as you actually change the goals change where they're written down it makes where everyone is on board of what we're doing next goals should not be like they should be constraining you and where you're going but they should not be limiting you overall design they should be a tool to help you rather than something to to stop you from getting to where you want to go so let's set some goals yeah like there's kind of three stages to our to our process I guess the first is this second goal stage where we sketch out the outline of what we want to do second step is going to be talking about relationships which is filling in the details and then in step 3 is the math which is basically doing the coloring in so I said we're gonna we're gonna go through a sample economy set up and I'm gonna lay out some goals and then we're gonna we're gonna go over them and figure out how to get that so go on incentivize the best parts of the

game this is the theory I was talking about earlier we want to make sure that the the most rewarding parts of our game where the bits players will enjoy the best it's a good starting goal nice and straightforward goal two we for our game want to allow as much as possible everyone to compete we don't want to have huge gaps between like the veteran players and the newer players we want to try as much as we can within the context of everything else working to allow everyone to have a fighting chance in every battle essentially so that's going to be driving some of our thinking for reasons that I will get into more as we go through this we want to maintain fairly stable price ranges we want fluctuations in prices where we want a player driven economy we want the market to be uncertain but we're going to want to to keep it within a certain range and not let things drift too far out of bounds so that we're not basically violating our balance assumption on top of those three this is kind of

order this is not a strict ordering but it's a general sense of the priorities we want to maximize the enjoyment of the non core audience which basically we know from experience from our game and from other sandbox games that some players are not gonna want to do PvP for example and even if we think it's that we think it's the best thing since sliced bread some players are just like I really want to play your game I just want to play in this safe area on my own that's okay we need to make sure that our really good bits are be all of us as guild and so on oh I did like as good as they can be but then we want to try and like make everyone else happy as much as we can on top of that so that's kind of a secondary consideration that's always in our head how can we have we fix things for them as well and finally we're gonna try as hard as we can to keep the math simple there's a place for complex math but it's a fairly rare place so as much as we can we're going to try and keep it keep it straightforward it means that it's easier for us as developers to work with all the numbers we're doing so making a lot of numbers and it means the

players kind of the simpler and more easy to grasp the math is the easier it is for them to develop an intuitive understanding of how the game fits together without having to go and build their own spreadsheet to figure out what you were doing so those are kind of our our core goals step one do the outline next step is step two building relationships filling that outline in so go on incentivize the best bits of the game we think that the best parts of our game the most intrinsically rewarding is our PvP and especially our guild vs. guild like territory warfare gameplay they score really high in our intrinsic motivators you're out there with your guild fighting other guilds who you know and you're you know your next door enemies who you love to hate so you've got a lot of relatedness your guild has control over where it's going and who it's fighting you get a lot of autonomy out of that and because we're doing PvP and we have this complex you are what you wear like equipment metagame each

battle can be different so we can have strong intrinsic rewards as well as so strong mastery because we're constantly facing new challenges to come over to overcome so we think that's the best bit of our game and we want to make sure that people go there so we have eight years of resource in our game and we have five resources the first three tiers are effectively training resources we don't really worry about them from a economic economy balance point of view the game kind of starts at T force we have T 4805 kind of crime we tiers that we're worried about our world is divided into clusters which is like a region a zone of the game they're like 800 meters aside I think or something squares the the makeup of the world each cluster has a tear tear at the cluster determines the here the resources present so at tier six cluster house was up to T 6 so

when we set our world up and you can't really see this but it's there I promise like in the Outlands in the PP area where the highest tier clusters are and and very kind of most competitive areas where the very highest years are so the TA clusters are where the kind of the core the guild vs. guild happens so the places where all that fighting happened is all that good gameplay happens also have the best resources and therefore the best rewards cool so that's kind of our our core thing you get the water for doing things that we think you'll enjoy second goal allowing everyone to compete what we're gonna aim for here in simple terms is a low power gap and a high rarity scaling what do I mean by this so we have a power progression between tiers you can have t for equipment t5 t6 t7 t8 each tier is more power from the last and we want to get that gap as low as possible so that the t8 people aren't so far ahead of the t4 people so we look

to other games we look to our own game we play test if we talk to players we've worked through it all and we figured out eventually that about 1.2 per level but here is about what we're looking for that's about as low as we can go so t5 is 20% more powerful than t4 t6 is 20% more powerful five and there 4080 is about twice as powerful as t4 2.07 so 1t a guy has about the same stats as 2t four guys that's about as low as we can make that without the tear step feeling empty and like what's the point in getting IT gear and that's kind of where we hit with that by contrast rarity of resources in the world we're gonna go much steeper we're gonna do 3 times per tier so with each additional tier is 3 times rather than the last so t8 is 81 times rarer than t4 what we're trying to do here is basically make the very high end stuff incredibly expensive kind of the ideal balancing

fall out of this setup is that t4 is common if you want to go PvP you can always afford some t4 gear it's not completely valueless because we want death to have some meaning it makes the fights more interesting but it's it's accessible t5 and t6 is kind of what you work for an everyday basis that's where you want to be hitting for most of your fights and then t7 and especially t8 material right at the top are very rare we want the big gills to be saving up there t8 equipment for critical fights if most players never see t8 at least for the first you know year of their game or whatever that's fine with us that's in us in a healthy balanced place we want this because the high-end gear needs to exist to act as a wealth sink at the high end because there can be otherwise you can get massive wealth disparities between different levels of player and also to give them a little long-term goal you're always working towards something once you get your

final goal you're kind of done like we found an EVE that like the biggest shipping game is a Titan and it's 20 kilometers long and it takes six months build and it's worth like hundreds of real-world dollars whatever and like something like last time I'd look many years ago like half of all type and pilots who had unsubscribed from the game because they'd won right they've got the biggest thing game done so we want to keep that high-end as rare as we can and I say use it use it as a way to sync resources and sync wealth out of the world without being completely overpowering to everybody else so that works out nicely next thing we want to do we want to keep these as I said earlier keep these price ranges stable to a certain degree we want price flow obviously we have a player market open economy player driven economy we want prices to shift according to supply and demand this kind of balances out the income per hour that

people get in different professions because like time is essentially the ultimate resource for any game like this we want to ensure that people who are killing mobs to get silver and making the same kind of money as people gathering resources and if we have the market then that those two will balance out as people if it's not worth doing people if there's too much of something the value of that thing will go down people will stop doing collecting it so much the supply will drop the value go back up like it's a nice happy relationship but we want to keep the overall prices within a band this is good for a couple of reasons it means that people can't the supply and gills can't run away with wealth by monopolizing the high-end resources and basically forming a cartel or monopoly there that can cause problems because they get so much wealth that it distorts the entire game we also have this thing obviously where we have these five versus five fights you can't just bring

more people to deal with the wealth imbalance so we're a little bit more constrained in terms of what we're okay with with disparities there and finally of course we need to make sure the board scale that those fighting areas the gbg the PvP areas vs. guild player versus player are more valuable than the safe space if that gets out of whack then our whole game swings out of truth so we're gonna try and keep those under control how we're gonna do that is we're gonna have overflows and under flows for everything so we have as I said we have our resources we have our silver as kind of the two things we need to balance off against each other to a large degree we have a couple of features we have this salvage feature that lets take a crafted item and salvage it and that get see you 75% of its resources back its resources and 25% of its resource value back as silver this means that players can turn resources into silver through this method at a rate that we set we also a

feature called transmutation you can take some tier forward to the transmuter here's a building you give it a load of money and it may give you some t5 wood so you couldn't you can improve your resources by spending large amounts of silver on it this means that you can turn silver into resources so now we have this back and forth flow between the two at rates that we set which is important what this ends up doing is building a price channel for these things if you're selling resources for less than its effective silver value then some player will figure that out and they will buy the resources craft them up Salvage them out turn them into silver for more silver than maybe than they bought them for and that will bring the price back up again as the demand shoots up and everything because people getting free money that gives you an effective floor on the price and similarly transmutation if you're selling above the transmute price then someone can just you know you're selling t6 at above the transmute price someone can buy some t5 transmute it up and sell

it and then put on the market for profit essentially free money again that will tend to drive the prices back down so we end up with this nice channel between salvage and transmute so that gives us our under flows and all other flows give us our price controls but as we're building relationships what we need to do is we need to find one that relationship is we assume that the sort of expected price based on rarity follows the rarity curve so the sort of the natural price for each resource goes up by a factor of three each tier we get our salvage price it gives us a nice floor and I got a trance peon price which gives us our ceiling this is a log scale because it flattens it out so we have this price channel and then we need to decide how that relates to the rarity progression because that's kind of the core like your peers how do those two things map together so we decide that a low end at t4 we want to have the salvage prize to be about the implied rarity value is we want to

ensure that those prices are T for low but we want to hold them up a bit we don't want T for equipment to be worthless because that makes dying worthless which makes it boring we also want to ensure that new players coming up they just get into Tier four for the first time they're harvesting Tier four resources we don't want that to be worthless so this kind of generates a floor worst-case scenario they can salvage it out and make silver off it so harvesting t4 is never not worth anything the transmit price we want high at this point because we don't want a veteran players bypassing those new players and just you know transmitting all the t4 they need themselves we want to ensure that the it's not worth transmuting down here basically by contrast at the high end we won our a tier eight we want the of implied gravity price actually even slightly above the transmute price so that we're ensuring that as much value as possible is coming out of that while keeping it

under control because the the limit at the high end is respawn rates in the in the territory that gills control rather than harvest time they're basically waiting for the resources to come back so there's basically free resources it's always going to be worth it for them but we don't want that price to shoot through the roof and we don't want a both because we don't want those gills become fabulously wealthy and because we don't want to lock everyone else completely out of the highest tier gear so in fact would be happy if most of our tier 8 came from transmutation it means as a huge silver sink we're sinking a lot of silver out of the economy which keeps our inflation under control makes us happy means everyone can get it but it's expensive we often want people salvaging high tea equipment that would be crazy so we have these low and highs and then because you know one logger off we're just gonna say a rarity scaling base that he needs to hit like this so we have this addition to these numbers now we have this thing called item value IV which is kind of our arbitrary silver

equivalents for a resource and we want to say the item value which is basically the salvage value is about the same as the rarity value at T 4 and the transmute cost is gonna be ten times that so you're always going to have this this item value we just what Salvage returns and above that you have the transmute value and we know it's pinned the the two things in Salvage and the resource value intersects at t4 we want them to cross over at t8 so our item value progression is about one point six six and that is basically our economy setup there's a whole bunch of numbers they're going to follow along from this as we map out like how much of everything on all the respawn rates and so I'm not going to talk about in a second but our core economy setup is this slide this is the basic relationships that determine how everything fits together and we'll do the same process for silver and for fame our XP couldn't of XP and for farming and for everything else we'll go through this process and how much do we want and what are the how fast you progress how much Fame do you

get per hour how much is it worth how long does it take you to max things out all this kind of stuff but it's all just a set of relationships and core numbers that say this is how the game fits together and that's our relationship building so we have these goals we have incentivizing the best part of the game we're doing that through saying the the bits the gvg and the pvp let's add the best resources allowing everyone to compete by keeping that power gap low and keeping the rarity progression steep someone didn't listen to the phone warning we're maintaining stable price ranges by having these overflows and under flows ourselves we cannot transmute to keep things in the channel they can float around all they like inside that but we know that they're not gonna get too far out of bounds and then along the way maximizing enjoyment of an encore audience that's things like for example transmuting means that you can get to th from the safe zones you can only harvest up to t5 but you can save up the silver to keep transmuting up and up and up it will take you forever but it's there making sure that t4 and t5 are available

in the safe zones so that you can kind of you can kind of gear up to go do PvP before you have to go into the PvP areas all this kind of stuff is just you know ticking away in our mind how can we tune this to make them happy and the math is pretty simple as all geometric progressions because we don't find a need at this point to make it any more complex than that cool so let's do some coloring in doing the math we have our relationships we've got all these free-floating relationships we need to pick some anchors or least an anchor to kind of pin that down to some actual numbers for this what we decided to do we looked at saying okay the high end of normal PPP is tier six and the people we think they're playing at that level we reckon that's me this meant anything I'd died this often and we need about 10,000 tier six per day across all the resource types so that kind of gives us our anchor point we know then and if it's 10,000 T six because of our three times

rarity progression it's 30,000 t5 and 3,000 t7 3333 t7 and so on from that anchor point and our relationships it all flows out and we make a giant spreadsheet and we go away and we do Excel and this is what I keep doing the work right now the last couple of weeks is making more spreadsheets there's a lot of work involving this for us because we we need to map out everything in the game but they shouldn't be surprises we always check the numbers at the end of the process to make sure they don't look insane and sometimes the things we want to fiddle with a bit for player experience reasons like we've been thinking a lot recently about how long does it take between finding resources to harvest and can we kind of tune that to be more enjoyable without breaking or overall economy but for the most part we don't want to be fiddling if we find ourselves wanting having the urge to change a lot of numbers at the end of the spreadsheet process then we know we've probably done something wrong

in either our goals or relationships and we want to go back and change those and we'll usually have like the set up variables will be in the heap somewhere talk about that some more in a second so you can ask oh okay what if we change it from three power of three per level or three to whatever the level 23.5 does that give us more it won't okay we're working at that level of the numbers rather than tuning individual values because we want to have these clear relationships that are driving all our stuff so in the context of this in Excel I want to talk a little bit about about some CEL best practices that we've developed because I think it's cool I like Excel it's my friend the table feature is amazing who here knows what I'm talking about okay cool so some people don't this gif is gonna blow your minds come on so you can basically select a thing and then you click make it into a table and then it's a table and then you can add things you can refer to them using the column name which is wonderful and

it Auto extends and you can filter based on the thing in it oh it's so good this I learned this at sandbox and I'm just like this is amazing so this is definitely thing we use a lot if you the color-coding gives you all this stuff a particular little trick I picked up is the variables table if you make a table that is a single row and you call it VARs then you can rather than having to refer to you know dollar a dollar thirty two is your balancing number for this it's its VARs primary mult that's where it is because the columns are single cell so it returns a column of a cell rather than an array so we often have these little settings tables just let me put all the vibes in one place and it's really easy to understand brings me to my next point usability and maintainability of Excel we're getting deep into nerd country here I know but I think about this a lot because we we have big spreadsheets we have a master heat we've been using for like three or four years that has all the kind of core data in got very spin off sheets that we use and there's different people using them at different

times and you have to come back to things that you've done from six months ago and we update things for new math and all this kind of stuff and kind of keeping those spreadsheets fresh and usable and present is is kind of a challenge so we think a lot while we're doing this about sending them off about color coding so we can see can I get back to this we can see here like this the orange tables are static game data we're not touching green is our primary setup the green cells are the cells that you want to tweak most the red cells are output data that we're going to put into the game and so on so we can look at this sheet and say okay I want to fiddle with it I know I'm fiddling here and here because that's where the green stuff is all this kind of stuff I use the comments feature an excel a lot you can when cells leave notes yourself think this is why I did this gigantic formula because in six months time I come back to hang have no idea what it's doing and generally thinking about the robustness of formulae making us they can deal with adding and removing columns or changing

column names it's a whole bunch of failure to do with vlookup and matching and stuff that I might get into but like we think about this stuff a lot we wanna make sure that I Excel can survive being punched in the face by people who don't know what they're doing that's point two and point three I'm a strong advocate for automating all your exports there's obviously an Argentine made about the saves design a time for - everything manually but then you get into an argument with your programmers about whose time is worth more obviously the designers time is worth a lot but let's just sidestep that let's talk about bugs right if you put things in my hand I've done a lot of hand entry in my time eventually that fat-fingers something and you're gonna add or miss a digit somewhere and something buried deep in the bowels of your game is ten times more or less than it should be and if you're a small team like us he probably won't catch that if you've just done a thousand numbers by hand and then some players will and they'll destroy your economy before you catch in it's a complete pain to unpick if you've automated all your exports you've built

a nice mathy Excel table that's all formulas so there can't be any individual errors and then you push a button and it goes into your game then that kind of bug doesn't exist anymore the bug where you've done something wrong and everything in your game is a factor turnout does exist but you spot that because it's really obvious so like we've had our CTO made us a nice little Excel plug-in that you click and click a button and it writes it out to a two-hour XML format which is lovely so I strongly strongly strongly do the automation you can all the arguments about about efficiency of time and to say it's gonna stop you from breaking your game because I've seen it far too many times yeah and then we ship the thing and then we see what happens for a couple of weeks ins firstly like we used to calculate on Eve wood 10:00 p.m. we get 10,000 players online after downtime so when you get 10 thousand man-hours of QA in our first hour which we can't match pre-launch period that's too large a number on top of that at a more kind of detailed level and more subtle level if your game is simple

enough for you to predict how it's going to play out then it's also simple enough for your players to predict how it's going to play out and some of them are smarter than you and they will solve your game and put the answer on a wiki and then everybody knows the most efficient way to level up is killing wolves everyone is now killing wolves congratulations your game is boring so I'm of the opinion that your game should be sufficiently complex that you do not know how it is going to get used which is kind of a scary thought but it seems to be the only way to square that circle of keeping people engaged over long periods they need a game that they can't see the end of they can't see that the simple solution - which means that you're gonna have to you know you want to take a good stab at it and all this work we've done here is doing that which when I figure out where we think everything's gonna settle and we have these you know this is part of why we have these min and Max things these are overflows under flows the the price channel so that if we've got something horribly wrong the game still catches it

but there's no real substitute for putting it out there and seeing how it goes so you're gonna have to iterate on it you have to see how it performs then you're gonna have to make changes when you realize the mistakes that you've been ever typically made basically how we're doing this is we're doing some science we're going to make some mental models we're gonna make some predictions based on those models we think the economy is going to do something like this we're gonna put it out there we're gonna test those predictions and then we're going to use those measurements those tests to refine our model and have another go at making it better and there's a lot of things that play into this obviously we use analytics we use metrics they're great in particular for the maker step figuring out what's going on but we also involve our players a lot we talk to our players we have forums we have a hand-picked group of players called our round table because we're artheon who give us more director and more sort of focused feedback we have a disc or we have reddit you know bla bla bla bla

always when I talk to our players because a they will spot they will tell you where to go look in your data be they'll give you a lot of qualitative information and see they will help you why which is the real real crux the Hewitt's how do you build a causal understanding of what is going on in your economy so that you know that if we change X then Y will happen if you just rely on metrics which should great and all you know you get some data points you like oh okay okay okay this is probably doing this and if your game is really simple that's great and you want to maximize the the height of the line you just keep Ben right if you're famous complex than a model may tell you actually know it's going to do this right and if you try to use your metrics to min/max that it's gonna best case take you to that peak on the right I'm gonna stop and worst case you're gonna go over it you're gonna crash it with it so metrics good but metrics good in that place feeding into your models and not just being blindly trusted because your game is simple enough to be solved to solve your balanced with metrics it's

too simple and your players will fit solve it for you so always in this always looking our feedback always looking at our analytics I'm always trying to kind of wanting ship to go back and constantly working back towards those core goals to bring our economy into the place we want it to be so that kind of is going to wrap it up understand what you're doing and why you're doing it and say I'm a philosophy guy so that kind of works me but I think it's important generally set clear goals invest time early on in your process and setting the goals for what you want to achieve spend a lot of time thinking about the relationships between things that's the core of your economy set up is how things relate to one other what it's telling your players to do where the incentives are and where that's going to end people up your math should be unsurprising and uneventful do your X cells well make lovely spreadsheets with pretty colors these tables so awesome and yeah math it out if you need to make changes

go back and changing your your goals and your relationships during your final numbers and then we ship and we revise and we look at our analytics and we listen to our players and we figure out how to optimize the game it's never going to be perfect because if it's perfect then it's solvable if it's solvable then it's boring cool that does me yeah thank you we have people that microphone is wonderful hi hey so near the beginning of your talk you went over the difference between extrinsic and intrinsic rewards yes and you mentioned that one of your goals is to basically overlap them as much as you can there is a lot of research out there that to us if you give extrinsic rewards to behaviors that are intrinsically motivating that it actually reduces the intrinsic motivation most of that research is in the context of workspaces and like job

performance but I was curious if you've encountered that trend in your work and how you've dealt with it so the question is basically what about all that research that says the extrinsic ly rewarding people for intrinsic legal warding stuff makes them like it less more or less yeah there's the thing in ideal world in a simpler game we wouldn't need all these structures but I'm not sure how you avoid having extrinsic motivators in a game like this I'm not sure how you'd make an MMO scale game without any kind of like XP or money or anything which will be an interesting thought experiment but I haven't like I don't know where you go with that so given that we have them we might as well make them use them for good right I know I know what you're talking about I totally agree in a bad deal if we didn't have that we don't have achievements we don't have any kind of sticking an RPG system on everything but in the game where we kind of we

don't see the obvious way to avoid it we've just got to kind of make the best of it if that makes sense Thanks you're welcome I just had a quick question about inflation you said before that as far as I'm inflation there's that old security system that that's fine I was kind of more curious about your end gameplay end game players and inflation on a player level well just over time accumulating resources how you would deal with sinking them like gold sinks in well for instance okay so the question if I understand it is how do we how do we kind of ensure that individual players don't get runaway wealth that ruins everything for them good question because inflation has historically been a big issue in all kinds of like persistent world games a couple of things one is say to always have a higher goal that they're aiming for that they can throw money out is a nice way to get them to spend but the thing we found like so we found on eve my data is all two years old so my shaft Mia this is wrong now but money in the game went up over time and average money

on players went over time but market inflation was pretty much flat what we think is going on here is that the players particular players who generate lots of wealth who are into that kind of gameplay they see their wallet as their high score so they're they don't they don't female all that money so they can throw it at someone make you make their money folks say I've got three trillion it how much have you got and that's kind of their their motivation my assumption least their motivation for gathering all that money so it doesn't impact the gameplay nearly as much as you'd think because they don't want to spend it that makes sense thank you I'd love if anyone did some research on that and told me definitively because it's an interesting question yes very interesting question I'm Ted Castronovo I'm a you know virtual economist this was a wonderful a talk I'd like you to design the real world economy like a three thousand word rant on what the financial system is completely badly designed if you want it

well yeah can I get your carpet anyway a question what do you guys do about gold farming I just looked up the price of Albion silver and haha there it is yeah so what do you do for that third market so we have a system the same system that Eve has more or less where you can we have in-game silver and game gold and silver is the main game we got the resource main game currency in gold is bought with real world money you can buy we have a premium subscription that increases all your progression by 50% so you really want it and you can buy that with dollars or with gold so you can trade gold and dollars effect or silver and real world money effectively on the market that puts a limit on it or like dampens down the gold selling on top of that various enforcement programs I don't know the details of but I know people are doing it's just a scourge of everything ever we're not quite sure I'm not aware of a silver

bullet solution but yeah we have we have the gold thing so people can trade money for game time and so on and we kind of only just try to stamp on them wherever we find them sadness hi all right thanks a lot for the talk I really enjoyed it my question is about AI participants in the economy I don't know if Albion has these but I know that Eve does so I'm just curious what sort of goals or approaches you take on getting AI players to participate in such a way that doesn't cause players to either ignore or abuse them so you're talking about AI in terms of the player one bot so in terms of our intervention in the economy no like in player factions or NPCs we're gonna buy sell or generate pieces of the economy so I said again in-game factions or NPCs right okay so yeah yeah I was developer side AI rather than player side a I yeah okay so

question is how do we intervene basically in the economy in an automated way largely speaking we don't we have things like for example when when the game goes live the first time it takes several days of people to get enough food supply to supply the buildings need to make food so we have some like sell orders for food first week of the game after launch to kind of kickstart that but otherwise I don't think we have any artificial buy or sell orders in the game it's all entirely player driven they can bootstrap everything themselves beyond the food issues so that's not actually problem and yeah we we don't mess with that side of things very much at all like between say stone wood do you try and have them all balanced at the same level or do you want to specialize in to the different types if find the question correctly is do we

have an equal amount of each resource type in the world five minutes thank you yeah we we have yeah at least certainly the level tonight here we we do the math so we know how many resources of each there needs to be and then we tune a spawn rate of each individual resource type so the slightly fewer wood nodes in the world cause less forest therefore would note has a slightly shorter respawn time so yeah we try and keep that pretty flat across all things because the goal is like the the reefs or the equipment construction setup is designed to use everything equally obviously if players find for example the cloth armor is way better than leather armor and there's going to be a surge of demand on cloth and drop for leather but that's cool that leads to all kinds of interesting market effects that were very happy with thanks I just I just wanted to talk more on the are asked a question about the inflation did you consider a pure bartering system where there is only a player market if you if I want your swords you want chickenheads I give you five chicken heads for our sword rather than a silver gold coin currency uh-huh

so question is did we consider not having currents in just doing barter I don't think we did it's an entry there was actually interesting thing a couple of tests ago where players were everyone was very obsessed about being self-sufficient all the gills wanted to make all their own stuffs they weren't actually using silver at all they were all about house and then all that players to go out and gather resources we can build our own stuff which led to a really weird place where no one needed money which mess with our game quite a lot but broadly speaking the main problem we're trying to do that isn't it's a really interesting approach but the problem is that it really clogs up the market because if you have X you need to have and you must want Y that's being sold it means it's much hard to find a market match whereas having silver as an intermediary means that you can always offload stuff silvers basically a way of just or money and currency and generally just a way of time shifting trades right so it means that you can offload that sword for some silver now and I can turn

the silver into the eggs I need later rather than having to wait for a silver two eggs trade to show up or doing some weird thing or you know Trading V Amex sauce or whatever to try and get myself to what I want so it would be cool but I think it ends up cloaking the market up too much for it to be workable at least without doing a lot of thinking yes I don't think there's any more questions I think I can stop there that's wonderful yes so thanks for coming you've been a lovely audience women to fill out your your forms and stuff I'm going to keep talking so that guy can't clap no I'm done cool thank you [Applause]