Mansions of Madness 2nd Edition: Postmortem of an App-Integrated Board Game

[Music] yep thank you for joining me at the crack of 10:00 a.m. to talk about mansions of madness my name is Andrew Fisher I'm the Borden card game manager at Fantasy Flight Games for those of you who aren't aware of us Fantasy Flight Games is a hobby market board game company we operate out of Minnesota here in the States so I've been a designer for Fantasy Flight for about seven years or so I've worked on the Star Wars role-playing game I worked on the rune wars miniatures game the fallout board game among many others I was also the scenario designer on mansions of madness which is what I'll be talking about today I also quit before I started the talk just wanted to thank a few of my other team members who helped contribute to this talk the lead designer of the game Nicky Valens couldn't be here and also a

couple of our scenario designers and ongoing developers on the game Carlson Carson tell dunk and grace holding house as well as our primary programmer we had a few different programmers working on the game but our primary programmer Mark Jones helped put together this talk and he's in the audience today so I just want to thank them cuz they were a huge help in putting this all together so to start us off that's a good sign there we go all right mansions of madness a second edition this is a board game we released it about two years ago we are still releasing expansions and DLC for it today but I'll just be kind of talking about the core game development today and kind of the challenges at creating this kind of unique form of game that I'll talk about so quick before we get started I kind of wanted to do the the standard GDC survey of how many people have played this game but first in our department we were kind of talking about we were curious about the

attendance of board game design day so I just kind of wanted to get a sense of like how many of you work on digital games okay yes so like the vast majority of you how many of you work on tabletop game all right sweet represent yeah so and then finally I'll ask the standard question how many of you have played mansions of madness second edition alright so a good number of you well for those who haven't as it's kind of says in the title mansions is what we call an app integrated board game you'll be excused for maybe not quite knowing what that means because we kind of made it up so what does that mean well you might first be thinking about playing Settlers of Catan or Ticket to Ride or something on your phone but those are more like digital board games they're exclusive to that device you also might be thinking about supplemental apps for the board games you've played like using an app to help out with with your character and

betrayal at house on the hill or that cool voiceover app that comes with one night ultimate werewolf but those are just kind of supplemental you can play these games without these apps what our app integrated games are are a game system that requires both the board game elements and the digital elements to function ideally these are game systems that couldn't exist in the same form without both elements we want to try to create craft a unique experience that you couldn't get quite any other way so what Manchin's actually looks like in play is you each play as a investigator in 1920s arkham arkham is this new england city in this world where all of the writings of HP lovecraft are true and so you are looking into the you know the occult and the working of the Elder Gods and kind of investigating these haunted mansions represented by these boards here well I shouldn't say they're

usually mansions some of our expansions go to more exciting locales but usually you're exploring these different mansions that we lay out in different ways with these different modular map tiles you move around the map you roll dice to its figure out a few pet succeed or fail at different challenges you find items represented by cards and you make important decisions that changed the outcome a branching narrative each game of mansions takes about two hours ish to play depending on the scenario and each play session of it is a self-contained experience at the end you'll have a story with an ending this isn't really a campaign game this is kind of each time you play is a self-contained experience all of the narrative in this board game is run by the app the app tracks all of the narrative and enemies basically it's

kind of like if you're playing a role-playing game or you're these investigators the app is your game master it is taking care of all of those tasks for you usually we ship on a lot of different platforms it comes on phones tablets you can get it on Steam on the computer and a lot of players actually we've seen online we'll put it up to their TVs so the whole group can kind of read it the actual game flow changes based on how people are interacting with it some people will just have the phone and tablet and read it out loud to each others others will pass it around others will all read it together on a larger screen so that's kind of the flow of the game is you're reading from this app exploring these mansions on the table don't worry if you don't know the exact specifics of the game I'll kind of dive into and explain specifics as I use them as examples so today I'll be diving into the core developments of this game and then I'll be diving into the core development of this game and then kind of going into a

post-mortem kind of a standard structure of five things that went well for us five things that went poorly for us and kind of what we learned from each of those especially about integrating tabletop and digital games together in the same dev cycle I'm not necessarily going to cover them in that order I'm kind of gonna cover them chronologically so with that let's start at the beginning so f of G has a long history of making great board games we've been making board games for over 20 years we also have a media department who handles our all of the supplemental apps for our games like dice rollers and such and they've also in the last several years been making some very successful standalone games based on our board games these kind of fall into that first category that I talked about these are tend to be adaptations they're not quite just ports of our board games to the digital space they tend to be they kind of tend to add to the game and take the spirit of the game and put it onto a digital platform so we've had success in both in both

realms and then we actually have at this point when we started mansions of madness had already shipped one app integrated game which was the XCOM board game in this game you play as members of the XCOM organization coordinating to save the world from aliens the app in this kind of encompasses a single mechanic it shows you all the threats coming down to the world in real time and you know the four of you have to sit there and coordinate in real time answering these threats as the app spits them out at you and it can adjust the difficulty kind of based on how well or poorly you're doing as you react to this app in real time but mansions of madness was on a whole different scale from this game so this was a much larger undertaking the reason for this is in XCOM be split between kind of the date the tabletop side of things and the digital side of things was pretty unbalanced and XCOM the vast majority of

the game mechanics I could have gone on listing them we were all in the tabletop side of things whereas the digital side just mainly handled the real-time execution of the threats so it is really just like a single mechanic in the game run by the digital side obviously you know this project had its own travails but it was on a much smaller scale than Manchin's was going to be and what we wanted to see from mansions so mansions was a daunting task but we got started right and that brings me to our the first thing that went pretty well with our development which was that we prototyped early Nicky Valens and I worked pretty early on the game and Dovan pretty thoroughly on our various prototypes to kind of see what apps were capable of in the realm of board games and what players would well put up with luckily we weren't starting from scratch as you can see from the title of the game second edition we are based on a

first edition of the game it's pretty similar to the premise I already talked about but instead of being a co-op game where you're playing as investigators investigating this mystery it was what we call our 1/1 versus many games so one player was kind of playing as the all the bad guys basically being your GM this worked fine but it wasn't as he didn't quite fit the game experience we wanted out of mansions of madness mansions was all about this kind of spooky storytelling atmosphere and when you know mike is sitting across the table from you you know playing all his cultists it's just not quite as scary as if you know it's this mystery coming out of the unknown so the one versus many wasn't quite working for the aesthetic we want to get across four mansions additionally first-edition had an incredibly complex setup as you seated you set up the whole board ahead of time you seeded out all the different the narrative onto the board and one little screw-up could completely screw up your

story for the whole two-hour play session also it lacked a lot of discovery you know you can not read the cards ahead of time but in the end you're getting all the content in the box you can see everything you're going to encounter in this mission you set it up beforehand so it just didn't have that sense of discovery that we wanted out of this game where you're we wanted this game where you open a door and you have no idea what's gonna be on the other side but it gave us a good launching a start launching point for our prototyping we started by replicating first edition but adding an app kind of in place of that one player so at our very first step we started by simulating the app we created these flowcharts these state machines of the apps logic and we created massive complex ones to play out an entire two-hour scenario with players and then we would have the designer kind of pretend to be the app in a play test in fact we went so far as to like separate them from the tests to some degree usually like one of us would sit with

our back to to the testers and just read off our flowcharts and that way to disconnect ourselves from the infirm a on the tabletop and really get in a headspace of like what does this app know and what doesn't it know and how is the interface with the app and the players but in the end our app wasn't going to be speaking to them so the next step of testing after we kind of iterated on this approach was to move to a gray box app luckily both Nicky and I have some coding experience and so we were able to write some rudimentary scripts that kind of mimic the functionality we wanted to see in the app so we've created this gray box that could play through an entire scenario of mansions primarily Nicky created it but so we could code but I don't actually think there's anything horrible in this little screenshot I showed but you don't want to look too closely at our gray box it's it's pretty gross there's like go twos and other things that I'm pretty sure it would make mark over there cringe it was it was a bit of a

nightmare but it was functional and it gave us the testing we needed it allowed us to identify very early on before real development started what the strengths and weaknesses of the tabletop elements and the digital elements were and kind of identify those to help us in where we were designing so we identified several key strengths and weaknesses here I've just kind of showed some of the the very broad strengths we identified I'll dive into them a little deeper and some of the things that went well and poorly and development but in general tabletop games are really good at flexible effects and timings at interrupting at any point at doing kind of weird incomparable effects they also have a great tactical interface for interacting for people to collaborate on a game very very quickly and intuitively they also have that social dynamic and we can actually do a lot of design around social interactions at the table that you just can't quite mimic in the digital space I guess yet VR is you know doing some

exciting stuff there but and then on the digital side of things the digital games tend to be much better at real-time interaction as they can handle that constant coming and going whereas like when you're on the tabletop everybody's like wait stop what does this card do what does this wording mean and all the bat also digital game excel at hidden information something that boardgames can often struggle with and then lastly digital games can handle a massive amount more complexity than our tabletop games can so after learning all these lessons and doing all this prototyping we were ready to dive into developments proper and this is when our tabletop teams started working with the media team and naturally these two teams worked perfectly together and the production went off without a hitch so that's a lie it was a bit of a mess and that brings me to the first thing that went wrong and several other things that went wrong table top designers and

digital designers working together was probably the one of the biggest challenges of this product and we hit it right off the bat some of you may recognize these icons the left there is a waterfall technique management technique where everything is kind of in a line and the other ones agile so our tabletop the tabletop side of things buy the necessities of printing require a certain amount of just linear scheduling things need to be done by certain points to start being printed or produced or tooled or whatever else whereas our software team can be a lot more agile and so we take two very different approaches and these immediately started colliding these teams had never really worked together before and these workflows really really clashed compounding this this probably isn't super accurate Waterfall makes tabletop look way too organized this is probably more accurate

well our tabletop teams are small and one of the strengths of them is how small they are that they can just turn around and chat with each other they can innovate you know you can just grab a card change the text right away it's that kind of mentality that kind of leads and we allow our table top designers quite a bit of freedom so that they can kind of innovate and you know make the changes of the game that the game needs this chaos isn't suited super well to them collaborating with our programmers and so this was a a pretty big conflict for us additionally this was one of the larger teams our company had ever worked with we tend to keep fairly small teams for our tabletop development and so we were scaling to a team size working across departments I know these are challenges that a lot of you are probably pretty used to but we're pretty new to us and so we struggled with communication we you know just turning around and chatting in each other's cubes wasn't really a good strategy anymore we didn't have like a

centralized communication thing like a slack or anything yet we didn't have centralized documentation we didn't have a middleman between the teams and so because of this we were kind of relying on just word of mouth and obviously that's kind of hit a brick wall for us we eventually got over it kind of early in development too and improved but one of the biggest steps we took was actually came from the media team which was the implementation of our tools originally you know we were working with this grey box and it was a bit of a mess and so yes all right so the a big breakthrough came for us at the discovery of playmaker playmaker so we were using unity we decided to use unity pretty early on I think we used it for XCOM is that right mark yeah so we were using unity and but the problem we were

running into is our table top developers us we weren't great at coding and then we also wanted to put developers on this project who had never coded before at all this is gonna be a big challenge for us and a big challenge for communication and training and everything else but playmaker was something that we discovered oh that's right yep so we have our gray box it as I mentioned earlier it was a mess it wasn't very scalable it wasn't really easy for non-programmers to learn and it was just a mess to debug this line by line with go twos just broke constantly and was a mess so playmaker is a plugin for unity that allows for visual scripting allows for create the creation of state machines to track the lot the the flow of the the game and it allowed it for us allowed for our media team to create a encapsulated environment for our tabletop developers to work in so that they didn't have to learn all the

nitty-gritty of unity and could just focus on designing scenarios and focus on the mechanics of the game playmaker was also pretty useful for us because it was very modular whereas with some of our other projects we create very like specific infrastructures unit playmaker allowed us to build on it and actually we used it for our next to app integrated games that came after mansions of madness this is the road to legend app for our descent board game and then the legends the alliance app for our Imperial assault board game both of these used that playmaker back-end that we originally developed for mansions so this ended up being great it really helped with our communication and problems because we could train the developers to use this encapsulated section of the game and of the development tools and work within that and allow it us as developers to focus on mechanics which is where we hit our next roadblock so those of you who know FFG know that we like to support our

games we create a lot of content this is Arkham Horror a second edition which I with I believe all the expansions it's kind of spilling off the photo there but we release a lot of expansions for our games we're constantly supporting and updating and creating new and interesting ways to play these games we need to create just a ton of content and it needs to be novel and interesting for these tabletop players it needs to play to those tabletop strengths we talked about and needs to be compelling in person as you kind of play the game with each other additionally as we're working on all these expansions we're kind of looking ahead and our media departments our programmers are going to be working on other projects they can't afford to constantly be supporting and expanding all the different app integrated game lines we you know we work on right now we need for the majority of the work for this to be done by tabletop developers now this is a challenge with

these kind of the mechanics that table top developers rely on to create interesting tabletop experiences tend to have a couple think problems that clash with digital so as an example let me show you just the test for mansions of madness now a test is a moments that something challenges you in the game and you have to see if you succeed or fail it's a pretty simple thing all you really do is you roll some dice you count up how many of a certain type of symbol there are in the dice and you get to have a chance to convert some of a second type of symbol to the first type to improve your result and then you count them up and based on how many you get you know whether you succeed or fail it's a pretty straightforward process but upon going through all of the content we've made four mansions thus far and looking at all the different places in this relatively simple process that our content interfaces with these rules we get this now this is if this looks messy but from a tabletop

perspective this is pretty straightforward we have pretty solid timing rules that tell you exactly when different things come up and it creates some very interesting effects and relationships that we couldn't otherwise create obviously this can be a challenge for digital implementation because so for example if this test process is all in your app we now need a user interrupt point at every single one of these places the user needs to be able to specify if some content that we've created is interacting with that test at that point that's a huge challenge and you can see it in games like magic a pretty sophisticated timing structure I think their online implementation has done a really good job but they have this same problem where they interface at all sorts of complex timings with their rule structure additionally the other kind of effect that we rely on a lot for development is these kind of in-person social effects think we can do things tactical things

with elements at the table and of the componentry on the table that are interesting in the rules we can do interesting social things like this is an insanity that you can get in the game that causes you to not be able to speak like in real life for the rest of the game to your teammates obviously things like this are a little more difficult to develop with straight digital integration an example an example of this is this actually pretty unique game by Harebrained Schemes called golem Arcana so this is a miniatures game where you play you control these different golems on this grid and they fight and it uses this really interesting pen interface that links to your phone and scans each of the models and keeps track of them throughout the game it's pretty cool but as you can see there's not a lot of componentry which is a benefit but all of the game execution is happening on those apps so because of that anytime

they want to develop new content they need a programmer they need to script it into the game and they're constrained by what the what the digital systems can handle they they're they can't kind of explore that freedom that we normally can with the tabletop space so because of that the kind of lesson we learned from this challenge that we ran into was kind of how to split up divided mechanics between those two spaces of the game between that tabletop side and that digital this side we divided we chose mechanics and divided them up pretty specifically into each space and tried to be as uncompromising as possible about how often they crossed over between those two every time they crossed over it can create kind of a messy point for example in mansions one of the things we didn't do so well was fire fire is represented by these little tokens they go on the board if you walk in them they do you damage it's it's pretty straightforward but we wanted

things in the app to care about whether the mansion was on fire we wanted like monsters to care we wanted MIT those events to care we wanted the fire to spread and all of this was handled by the app and so because of this this simple little fire token was existing in both worlds and what ended up having to happen is if you go into the menu main menu of mansions while you're playing there's just a big old set fire button taking up one of our new slots because it was the only way we could get into the app that a fire had started through tabletop mechanics so as you can see that's a pretty edgy solution to this problem but it's it's immediately created by a mechanic kind of straddling both sides so when we have mechanics that stay completely encapsulated on one side we avoid kind of those interfacing problems which I'll talk about a bit later as well so the next thing so once we divided those mechanics between the tabletop and digital allowed it allowed us to start really leveraging the strengths of both

sides and one of the places that really succeeded for us was the ability to conceal complexity remember Arkham from earlier this mess our games get a little complex we like to create a lot of content and you know keep our players receiving fresh new writing all the time and telling fresh news stories tabletop games by necessity put this all upfront think about walking up to this as a new player you know you've just got to gone over your friends house and they're like no no it's gonna be easy I swear now behold it's it's pretty intimidating but mansions is different mansions has even more content than this in actuality however you don't see any of it in fact the players only know it really don't need to know three things to play the game which is our three actions they need to be able to know how to move around the board they need to be able to attack monsters and they need to need to be able to interact with tokens so

moving is pretty straightforward you just take your figurine you move it around the board so we don't really talk about that but these other two fairly simple actions that are just clicking on a monster and clicking on a point of interest in the app are both far more sophisticated than they originally seem to a brand new player our attacks they're based on the weapon type you're using they can consider all sorts of factors they pull from based on the monster you're attacking they can pull from story relevant information they can consider all sorts of factors that are going on in the game itself all with a single push of a button similarly once you interact with a point of interest in the story these are interacting with the state machines we put together with playmaker when you interact you're getting story writing you're making interesting choices about the branching narrative you're seeing the repercussions of your past choices you're exploring new rooms which place new tiles you're managing the branching narrative all of this is managed by the app all at the push of a

button with just a few selections by the players so it's taking this incredibly complex system and making it incredibly easy for new players to interface with and I think this approach has really shown in our reviews for those of you who aren't too prepared there's this website called BoardGameGeek it's basically the IMDB of board games it allows users to post all sorts of stuff it it has all the information you need on board games and one of the things that has is a complexity rating for each game this is created by the users now as you can see here our original complexity rating for first edition was 3.2 for and with second edition we reduced that by over 20% down to two point six five all while having over a dozen like over twelve times the content that mansion's first ed had in that core box but there's a downside that many of you probably have seen to this and that it's

we actually had to develop that complexity tabletop developers don't often get the chance to just run wild we're really limited by the budgets on our printed components you know we have X number of cards that we can afford to put in a box and that's it but now the sky was the limit and our eyes got a little bigger than our stomach in fact they got a lot bigger than our stomachs so here is a sample of what we originally planned to put in the game so we originally planned to have five of these branching scenarios for you to play each with what we for what we call alt maps which are basically kind of maps that we randomized some of our maps but these were more sophisticated rebuilds of the maps placing them in different look Asians and such so that gave you 20 different map significantly different maps to play on we had nine different monster types with five attack types

each we had six types of mythos events and we wanted like oh let's throw two dozen of each mythos type into these six types we had puzzles each with skins as you can see this is a pretty big list but then on top of this this game was coming out fairly soon after the first edition of our game and a lot of players were still playing that and it just bought it and we didn't want them to feel like that was invalidated so the decision was made to integrate and make compatible all of the first edition content which is awesome it's really exciting you can use all your miniatures but for us these numbers ballooned a bit we added three extra alt maps to each thing to use the first edition tiles we added all sorts of monsters we added them to whole new mythos a type to fit with these new types of tiles that were coming in this got really out of hand they've ballooned out of control we

ended up having to make a ton of cuts to the final game from this planned content the core game only had four scenarios in it with a lot fewer alt maps and a lot of these attack types and events and stuff we had to had reduced variability early on in the game because of how much bigger our eyes got than our stomachs luckily now that the games been out for several years and we have a lot of expansions we've been able to fill out a lot of that variety by patching stuff into the game but at launch we had to cut a lot of this content because of how much our development was ballooning out of control on top of all of this this is all being created by tabletop developers who as we mentioned aren't great at coding we we had a lot of bugs and our tabletop testers they're not software QA people they you know don't have the skills necessary to hunt down and squash all of our bugs so in fact our media people our small number of programmers ended up having to be the people who

were squashing a lot of the bugs created by tabletop people you know we tried to hunt down our own but we were only so good at it so poor mark over there had spent way too much time squashing our bugs these delays all of that development delay all of these bugs added up this project really went on a long time in fact it went on so long that road to legend which I mentioned earlier the the next game to use this are this back-end that we developed here actually came out before us so we we took a while so on top of all of this development ballooning out of control we discovered something on the mechanical side of things in the game systems which was our game had a serious lack of transparency what I mean by this was the players were having trouble knowing everything that was going on we created all this complexity we created all this interesting decision-making but our players weren't seeing it I think upon reflection I think a lot of

this problem arises from the mediums this kind of difference of tabletop and digital that we've been talking about tabletop games are transparent by default everything in a tabletop game when you take it out of box you can read it all you can see everything you have to understand basically every rule in the game to start playing obviously that's not true people play magic without reading the 140 page rulebook or whatever but you need to have a full understanding of the game to play for for tabletop games they start transparent every mechanic is default transparent until you deliberately hide it from the players and tell them to hide it meanwhile on the digital side of things everything starts opaque you have a black screen until you add graphics until you add UI elements until you add sound all anything you want your players to know about what's in the digital side of things you have to deliberately put into the game and so you know a lot of this as we talked about was being developed by tabletop developers and so they weren't used to having to put all

of this transparency into their game and so one of the issues we ran into with Manchin's is that our primary interface were these text boxes as you can see the UI is pretty sparse otherwise but we have a lot of sophisticated information about how these scenarios are running that we want to communicate to the player so all of that information ended up having to come across in these text boxes as a result we ended up spending a lot of time in these scenarios that we kind of wish we could have been telling stories we had to spend explaining to them how the game worked and explaining them to them how the game works again and again and again a lot of our scenarios get pretty bogged down with just mechanical text as a result we learned from this in our later iterations like here is a the Imperial assault app legends the Alliance we learned from it we put a lot of our information in front and center you have a lot more sophisticated information up front that allows you to kind of see what's going on and these are a lot of

things that in mansions we may have had to give you text notifications about but in mansions it was it was a pretty big challenge for us but it ended up being something we learned a lot about for making these kinds of games in the future which is that when you're separating the mechanics between the tabletop side and the digital side and being uncompromising about how often they crossover one of the great ways to decide beyond those strengths we talked about earlier to decide which side a mechanic would probably be strongest in is based on the opacity of that mechanic is this something that should be hidden to the player by default then it probably belongs in the digital side of your game it'll be a lot less work to work on that mechanic is this something that's transparent by default you might want it on the tabletop side of your game because the player is going to see all of it right at the start and once we got a handle on this difference and what it meant to us and on how opaque digital games can be this opacity was actually

super useful to us one of the biggest strengths of mansions in the end and one of the things fans love about it is the sense of discovery it gets as I mentioned everything in a tabletop game starts out very transparent to the player but we could hide things from the players we could truly surprise them we could allow them to interact with hidden information and really sophisticated ways and it also allowed for really strengthened replayability as you saw in the component we only had four scenarios so we wanted people to be able to play these a lot and in fact I'm gonna give an example from the game so minor spoilers for the very first scenario it's called cycle of eternity here are some of our unity analytics on cycle these are the number of people who have played each of our scenarios as you can see Co e over on the right that cycle of eternity it gets played a lot it's kind of our default

learning scenario so a lot of it's a lot of people's entry point to the game you know you played it a couple times but you have that friend over who has never played mansions before alright well we'll play cycle again because cycle is a good intro and so game groups will play cycle over and over again because it as they bring new people into the game so it's also our learning scenario and so it's a lot simpler and shorter than a lot of others so it lacked that variance that a lot of other scenarios get so in it we actually got a really cool opportunity to kind of surprise people a bit so actually the the picture I've been showing is in from cycle of eternity this is the Vanderbilt estate you are investigating the mysterious William Vanderbilt you arrive at the mansion and you meet his befuddled but kindly Butler Eugene he's worried that something's going on with his master and the spoiler has William Vanderbilt occultists he's worshiping an elder God because everybody does in this setting

and he's trying to open a portal or something evil so that's how the game played out most the time you help you Eugene helps you you find Willy William Vanderbilt and you foil his plans but we realized that we kind of wanted to surprise these players were playing the scenario so many times so we added a a random role at the beginning of the game that determined a kind of Easter egg variance maybe like 1 in every 10 times 2 seconds we added this random variant in where 1 out of every 10 times you get to the big finale and William Vanderbilt is completely innocent and it turns out that his Butler Eugene has been evil the entire time and so this was a really awesome cute surprise for players people would talk about it on our forums talk about no II totally Eugene was evil and people would be like I've played this scenario like 10 times and Eugene was never evil you're crazy this is something you know

it's a cute little thing you know a lot of digital games could do it really easily but in tabletop it's it's nearly impossible to surprise players like this and to truly give them that sense this is one example but the game is full of this little sense of discovery we have maps where we keep most of the map the same so you think you've seen it before and then you open a door in a rooms difference and there's all these little things that we can kind of hide from the players and and kind of give them these surprises additionally as I mentioned this game branches a lot and this also gives kind of players the ability to kind of discover what we've created this is obviously an oversimplification but our main plots will kind of branch out from different nodes to different possibilities ending in different things so this gives you some replayability you know you can play through the scenario a couple times before you feel like you've seen it all but then in every scenario in addition to these main plots we try to kind of weave in various side plots

that are entirely independent and can kind of go one way or the other and so ideally by the time you multiply out three or four side plots in combination with the main plot going on with you know four different players playing investigators they've ever played before you kind of get this interesting emergent combination that you've never seen before and so it allows for each scenario to play very differently and it's something that we've been pretty happy with additionally we added on an extra layer of randomness onto this to kind of give you that variance play to play which was our mythos events I touch them a little earlier but basically they're just random horrible things that can happen to the investigators as they play and as you saw we developed a lot of them and now there's really a lot of them we have hundreds of these things and we put all sorts of different writers on we grab people from throughout the studio to write mythos events they're actually quite fun and these mythic and those events they pull from this random this

massive list but we tie specific conditions to each one so things like what rooms are in things that the app knows what rooms are in the game right now what choices of the players made what monsters are on the table what items are on the table we also have this threat system which is kind of like the director from left 4 dead in the sense that it knows how well or poorly the players are doing and can kind of ramp up for a ramp down the threat to kind of keep the pacing interesting for them mostly using these mythos events and so we we developed this system and this is this is also worked out great for us but it caused us to make an interesting discovery which is that we don't really see in tabletop games that often which is players and players create intent out of randomness we'd spent all this time linking these events to like you know make them feel like they had meaning in a scenario but players will ascribe meaning to the most random things in

tabletop they always know that's random you know they roll the die they draw the card they know that that was completely random chance and that nothing else fed into it but in the digital side because of that assumed opacity the players imaginations run wild they they think that that mythos event coming up means something in the story they think that oh this is based on my choice from earlier they create all this meaning that we just it's nearly impossible for us to create in a tabletop space because of how clear that is to the players and I think my favorite example of kind of using the players hypothesizing is so when the director when our threat system is low and it's it's giving them kind of a reprieve will give them some flavor text and will say no it like you know the wind whistles through the through the mansion creeks no effect right this was fine players knew no effect okay we got a break for the round but that didn't quite work for our Lovecraft II horror theme so with one simple word we changed

no effect to no immediate effect this was amazing players get so mad about no immediate effect they they they hypothesize about what that could mean that they you know like when they make a choice and it just pops out no immediate effect they're like what what's gonna happen this added to that sense a sense of dread that we wanted in a Lovecraft game and so we kind of played to that strength of the digital side and that opacity and that turned out great for us another example of this to kind of go to a couple another app integrated game that uses hidden information in the digital side really well is this game called alchemists this is kind of a cool game so it's a deduction game like clue but instead of there being like a murder mystery there are alchemical formulas

that you need to know there are all sorts of ingredients in the game you can kind of see them that the cards perched near the top of the little dividers and each one has like elemental principles tied to them and based on how those elements combine they form potions but you don't know which elements are tied to which components every time you play and so you just have to experiment by mashing these components together and making potions and maybe blowing yourself up until you figure out the deduction puzzle this game is cool because it's app does something that clue never could which is inclu the moment you check the file to see if you're right the game is over for you you've seen that hidden information and you've ruined the game for you but the app allows us to check against that hidden information in really sophisticated ways so an alchemists it can constantly every time you use that you use those components it's checking against that hidden information and giving you an output based on that that's something that a tabletop game can't do without a dedicated player

who's basically playing the system and that's how alchemist gets around it if you don't have an app is they have a role for somebody who just sits there and tells you what happens with your potions but the app allows that player to be able to play the game instead which is it which is really nice so back to mansions so we you know we went through these trials and travail and finally after all this hard work we're getting to the end of the game we're reaching the end of development which for tabletop games still means there's like nine months before it actually is on store shelves but we're reaching the end at it's time to announce the game and unfortunately this was another road bump for it which was our announcement was a little rocky public reaction wasn't great when we first announced it there was a lot of skepticism about the app and grated format we saw a little bit of XCOM as well but when we were taking some first ed game that was exclusively that was

exclusively tabletop and making it a Pinta grated there's a lot of resistance to this people were concerned about will my platform be supported will I be able to play it on X or Y will this game be able to be playable in 20 years with my grandchildren or whatever else you know I won't be able to play this board game after the nuclear apocalypse or whatever else this app is going to be a hassle so there was a lot of skepticism but we were really confident in the product we really liked the game and so we were confident that when it launched people would like the game but then our launch came and our launch was really buggy all those bugs I mentioned earlier it turns out we didn't manage to squash them all and we had a lot of stability issues at launch you can see here this is a super unofficial poll from BoardGameGeek on certain platforms certain scenarios crashed fatally like a third of the time

we had a bunch of a note and this is on top of a bunch of annoying small bugs that were also plaguing users constantly to our programmers credit they got on it super fast and like after a week or two a panic we had gotten the build a lot more stable but this kind of strengthened those original skepticisms about the app integrated format and that was that was a pretty rocky place for us I think the main takeaway from a lot reading a lot of people's skepticism though is ties in my last point which is why is the app worth the hassle this sentiment is something that we saw kind of over and over again like why not just make this a tabletop game why not don't just make this a digital game why go through this pain in the ass to combine these two formats when I could just get one or the other and so I think our last lesson and

something that went really right with the game is knowing the answer to this question and that the app needs to be justified its inclusion needs to be justified I think I've talked a bit about want some of the things that the app has done for us we feel like it's worth it right so what's the last element here well what is the hassle of the app right some people might say oh you know it's it's pretty easy to use there is no hassle right but let's talk a bit about the hassle and minimizing it because I think it's something we did pretty well so this is something I think I just kind of started using this term interfacing but basically this is the concept of any time your system has to cross over from the tabletop to the app or from the app to the tabletop anytime those two sides of the system need to interface with each other this can be done in myriad ways but you know like with audio cues with text with tapping with camera some much more expensive than others

but this needs be done in some way because we've so strictly divided our mechanics between the two sides of the system and they're not aware of each other so in mansions the mansion the app it knows your story really well it knows where you're at but it has no idea where the characters are on the map that's entirely on the tabletop it doesn't know where features are it doesn't know where the items have ended up it doesn't know where the monsters are all of that gameplay is happening on your table the app is just controlling a lot of the other aspects and so anytime the app needs to know that you need to be inputting into the app and also you don't know the story anytime the app needs to tell you story or where to place tiles it's outputting that information to you and the problem with this is that any time a player spends interfacing with the app at the table with other people is a time away from the game yeah I like

that figure that's picture for net runner all that time interfacing to tracks from that tabletop experience it's viewed by the players as work and hassle every moment that you know your friend over there is sitting with their face in the phone waiting to see what will come out of it is just kind of viewed as wasted time now there's many innovative ways kind of around this you know alchemists that I showed earlier uses image recognition to view the the components this game here world of yo-ho by yellow is pretty cool the the game mat has spaces that are roughly the size of phones and the phones actually show the board through them and you place them on the map and actually it works pretty well they use these the accelerometer in the phone to track where you've moved it around the map so it's constantly showing the right part of the map and it can kind of reveal encounters and stuff so it kind of gets that interfacing out of the way by directly integrating the phone into the game which is pretty neat we didn't

really have that option the stories we want to tell are pretty sophisticated so we rely a lot on text we have some narration but we couldn't afford to narrate all of those mount that mountain of content we wanted to do and so the question the question that helped us the most when we were kind of working on this as we're playing the game asking yourself am i doing the work for the app or is the app doing the work for me this is this gets down to the heart of the hassle do they feel like interacting with that app is a drain on their time so when you're inputting when you're doing that kind of stuff it's going to be viewed as you doing work for the app whereas like when you're getting something out of the app suddenly it feels like this is a benefit a boon to the game here's an example wrote a legend which was the follow-up to mansions based on descent because it was

based on descent each player activating mattered and when they activated mattered in descent and so because of that you can see over on the left there every time you activated you needed to click and turn on your own turn for all four players every round seems simple it's like two taps right but this really drained on the players they have to remember to do this they have to keep that in their mind they had to be passing around at the phone to do this on the phone this this is kind of like that player at the table you need to constantly be reminding to take their turn you're constantly reminding that app to take its turn and that would that was view this real drain on the players contrast that with mansions every time you're clicking on one of these nodes we're giving you story we're giving you rewards we're giving you choices we're giving you stuff players never responded negatively it's interacting with nodes on the app they were excited they wanted to see what was coming next so the way we but the problem is as I mentioned if

the mansions app doesn't know a lot of things about what the players are doing so what we tried to do is we tried to take this rewarding experience of getting something out of the app and harvest as much information from it as we possibly could so when players are interacting with the app and getting stuff out we try to look for every piece of information this gives us when somebody interacts with an app we know that investigator is on that space because they have to be in that space to interact with it we know when they interacted with it they know how we know how well they did we know the amount of damage we've given out to them we know the amount that how close they are to losing the game so we kind of use these things that players view is beneficial to get just enough information to make the app seem informed about what's going on on the table but I think there's a long way to go in interfacing I think we've succeeded pretty well in mansions but I think it's one of the places that these kind of hybrid games can innovate the most image recognition and other

sects that such technologies are going to allow us to kind of bridge this interfacing gap and make this feel like less of a hassle to players in the end I think mansions went pretty well despite it may sound like gloom and doom I talked about the rocky reception but actually after that initial bump we got those bugs fixed and the game got great reviews it's it's quite beloved and it's been going on with its expansions now for several years um and it's doing incredibly well for us so I think the big takeaway is from mansions development were obviously to prototype early to prototype thoroughly before you start development use that prototyping to be wary of scope and to plan your scope carefully and then with when you're developing for both tabletop and digital simultaneously consider how you're going to divide the mechanics between the two mediums and be pretty strict about that division consider the

flexibility of your timing versus the desire for real-time interaction um consider the opacity of the mechanic and how it's used by the players consider the interfacing and like if you want it to be shared and tactical and do you want to have social elements in this market for a given mechanic before you decide which side to put it on and finally make each interface with the app worth the hassle for the players so that's it thank you very much remember to fill out your evaluations and I think I have time for like a question or two oh yeah walk up to the mic hey great job I just wanted to ask do you have any

thoughts like on specific game mechanics that could leverage you know augmented reality things like a our kid in a Ark or for tabletop games oh yeah I mean like AR is something that we've talked about a lot it could be really cool one of the things I think we think could be coolest about it is the ability that kind of hidden information and overlapping asymmetrical information for different players sitting around the table so AR would allow an overlay of information over a board game so that different players knew different things about what's going on there I and like gave you that really open interface to the game through through AR while also maintaining that that really strong asymmetry obviously this is something the technology is a little far out for us to work on it or anything but I think it could be the potential there's really cool thanks a lot yeah I have a question

like you mentioned this this interfacing issue between the distance - it was made between the app and players have you ever thought while developing mentions of malice about reading out the text loud I mean the using any voice over or kind of this kind of possibility to make the game better yeah definitely um oh I just remembered I'm supposed to repeat the questions after people say them I I forgot to do that for the first one yeah so the question was have we considered doing voiceovers or narration so we originally had wanted to do narration for nearly all the texts obviously that proved prohibitively expensive and you know we talked a bit internally about like oh it'd be cool to like Microsoft Sam this all this text but one of the things were really going for with Manchin's is the atmosphere and at the time there was no real like

automated reading of that text that kind of got across Lovecraftian feel that we were really happy with so we opted to just have the players read it at the table also often players reading the text together at the table can kind of be this bonding and joking experience that has actually added to the game quite a bit so we probably have no intention to put anything like that in the game I imagine thank you a great talk and just wondering when you were there whelping the app and the game were you considering any sort of accessibility issues like in terms of the text and the speed of the text and the cars and these other things yeah so accessibility definitely came up oh yeah so the question was did we consider accessibility while developing this app and yeah accessibility definitely came up it is something that my departments in the tape in board and card is very concerned about and and we wanted to help people with we definitely had to

make some sacrifices just to get the app working it was something we're pretty inexperienced with but like app contrast scalability and stuff is definitely something we considered during development did you look at those old games from the late seventies in the early eighties where the limit technical limitations were so big like Odyssey 2 for examples so that you had those spot game things to to broaden the scope of the of the video game do - look at those interactions between analog and digital yeah so the question was did we look at those old games from like the 70s in the 80s that use kind of the crossover of technology yes so we what's the name of that one I'm blanking on the name but we we huh oh yeah yeah we we brought a few into the office and we're playing them it was it was it was a lot

of fun a couple of them had been sitting in people's basements and didn't really work that well but we definitely checked out several of those and kind of talked about that another kind of inspiration for this kind of game was like old school like adventure games and like like kind of like those FMV mysteries and stuff too so alright cool so I think that we have to wrap that up there but if anybody else has any other questions I'll be in the wrap-up area for a little bit so feel free to come by and talk so thank you everyone [Applause]