One More Run: The Making of 'Spelunky 2'

[Music] hi i'm derek you and thanks for watching my talk one more run the making of spelunky 2. here's a brief outline of the talk we're going to talk about roguelikes why i wanted to make a sequel to spelunky the challenges of making it the spiky design of spelunky what i was trying to accomplish the release my takeaways and then at the end i'll briefly talk about whether i can imagine more spelunky in the future or if this is it all right to start let me show you a quick timeline of the spelunky series which started as a freeware game i made in game maker in 2009 in red are the first releases of each game plus the book i wrote about spelunky's development i wanted to make a game that had the complexity and player agency of a roguelike but could be accessed as

simply as a platformer and i was lucky the concept took off spelunky inspired some awesome games like the binding of isaac and ftl and the rest is history even in 2021 the genre has showed no signs of slowing down since there was a release of returnal earlier this year as well as a big expansion to binding of isaac so why did roblox why did roguelikes take off and why are they still so popular i think the most obvious reason is that they offer a lot of replayability due to the random level generation they're fun to play and they're fun to make the results of random level generation can be as surprising for the developer as they are for players and they offer great streaming and youtube content so that that seems simple enough right but i think another reason why roguelikes are still going is that they embody a lot of the exciting potential

of the future of video games it's not just replayability replayability is really just a side effect of a bigger concept which is the exploration of large detailed and highly interactive worlds video games are an interactive medium that's what separates them from other art forms so randomness offers replayability sure that's the basic idea behind slot machines for example but in roguelikes i think it adds much more it's a way to make an infinitely large world within the context of traditional run-based gameplay even if an individual run plays more like a retro title across many many runs the structure starts to feel more like an open world game that's why the traditional turn-based roguelikes have remained such a strong niche despite the ascii graphics and obtuse interfaces

and that's why modern roguelikes exploded once those graphical and ui barriers were removed finally i think roguelikes are so popular because it's a flexible framework that happened to be tied to a very specific type of game in the past the berlin interpretation created for classic roguelikes is just that it's an interpretation there are high value factors and low value factors but it's not super strict it's not even canon classic roguelike devs have been arguing about this stuff long before the modern roguelite came along these factors are in effect ingredients that you can sprinkle onto your game to make it more or less roguelike what splunky did was simply show clearly that a lot of what was interesting about roguelikes were not actually tied to the turn-based dungeon crawl it definitely wasn't the first action rogue-like but it made it clear how flexible these roguelike qualities

really were and i've said this before but i think roguelike is similar to rpg in that sense any game can have rpg elements in any game can have some roguelike elements and that flexibility is a big reason why it's really just a powerful collection of ideas so that's why roguelikes are so still so popular why make a sequel to spelunky though well i had a few reasons as you can see here the first reason i wanted to make another splunky game was i felt like it still had more potential when i made splunky classic it took most of my focus just to bring these two ideas together roguelikes and platformers uh there wasn't a blueprint for that type of game yet when i started working on it and i was also learning game maker and doing my own programming for the first time in years so there was just a lot to learn but once spelunky classic was done it became it became the blueprint for splunky1 and with the help of andy hull eric

circa uh and microsoft we were able to to do a lot more to get closer to the edges of the possibility space of this world and these systems it also felt like we were getting closer to the soul of spelunky with spelunky one and i wanted to take another step in that direction if you think about how you know mario started with donkey kong and how each new game in the series moved it closer to where it is now with new characters new mechanics more polish that's kind of what i wanted to do with spelunky so splunky classic was donkey kong and splunky1 was like super mario brothers and you know essentially i wanted to make the equivalent of a super mario world reason number two uh as to why i want to make a sequel as i said before roguelike is a flexible framework and after spelunky new

roguelike games moved in different directions really quickly which is you know i think is awesome it just makes the genre just feel very diverse and interesting for example the binding of isaac is a top-down shooter ftl is a real-time tactics game set in space eventually slay the spire would incorporate roguelike elements into a deck building game creating another new subgenre so it's a really wild like explosion in every direction after spelunky and years after splunky was released roguelikes are still i think really popular but i still felt like spelunky was a relatively unique game um compared to all the other roguelikes i think the main thing that makes splonky unique to me is the way that it combines the simplicity of playing an action arcade game with what i like to call high interactivity so what is high interactivity

i think it's a combination of low level similarities in the berlin interpretation they call this monsters are similar to players and they call it a low value factor but for me it's one of the highest value factors that was really ahead of its time and then there's destructible terrain it's rare to play games that have fully destructible terrain but it's one of my favorite elements in any game that has it i think there's a ton of potential and destructible terrain for the future because it's realistic and because it also leads to creative solutions which to me are always more compelling than singular designer invented solutions and then finally surprising combinations like the rube goldberg esque deaths in spelunky that that always feel kind of comically tragic so high interactivity was one of those ideas from roguelike likes that i was personally most excited to bring over to

platformers to be able to access even a fraction of say nethack's interactive complexity through the simple controls and fast-paced action of a platformer to me that was so exciting and i didn't feel like a lot of uh like a lot of roguelikes really leaned into it in exactly the same way that spelunky did um which i think is fine they explored other cool ideas but you know if other developers had made games that just that did what spelunky did but better i'd say all right great work i'm gonna move on to something else so in the end for me it came down to potential the potential of roguelikes as a framework and the potential of spelunky as uh another kind of framework but the question was how how do we realize that potential to make spelunky one andy eric and i pretty much maxed out what we could do in a few years even with microsoft support

and besides andy was busy with his new project dunk lord so so he wouldn't be around to work on the sequel which brings us to reason number three why i wanted to make splunky 2 which is the right team i just wanted uh just wanting to make something rather isn't enough to make it worth making in my opinion from like a practical point of view in terms of commercial game development especially if it's a project as big and personal as splunky 2. and a lot of the new features that i wanted to add were technically challenging like online multiplayer and fluid dynamics thankfully bloodworks really seemed like the perfect studio to partner with we were introduced to them by sony in 2013 to help us port splunky on a playstation it was a great experience they definitely had the engineering experience and just as importantly it was it was clear that we worked together

well they were also one of the few groups of people that uh you know by by the time i wanted to make splunky 2 they understood how spelunky worked and how it was put together as well as me and andy uh and on the audio side i asked eric if he'd be interested in working on spocky2 and he was so we ended up with a great team to make spelunky too and of course you know we all missed andy and would have loved to work with him again but he was working on something cool too at the time and you know we definitely kept in touch during development actually it was andy who suggested adding the back layer to spelunky 2 which really gave the level design an extra dimension but we'll get to that later in the talk and finally reason number four why i wanted to make splunky 2 it may seem strange now but back then there weren't a lot of indie sequels being made it felt like every

time an indie game developer had a hit on their hands they were just totally burnt out on the idea afterwards and wanted to try something completely different so part of me wanted to make a sequel just to you know be be a little different and also to challenge myself and see what it was like which as an artist is always a fair enough reason to do something i think so here's the core development team with the two derricks me and derek leo our trailer maker in the us justin chan our illustrator in canada blit and our testing team at lollipop robot in spain and eric in norway once the team was assembled the project moved very quickly since blitworks had the code for the original game they started by essentially rebuilding a skeletal version of spelunky with new code that would enable online multiplayer and other features we might need including debugging tools that the

original game never had with spelunky one andy was always lamenting our lack of tools partly because tools are fun to make and i think he really wanted to make them but i got pretty comfortable with our janky process of making spelunky one and i was also worried about going down a rabbit hole of tool building that would delay the game even further because not only do the tools take time to write but a lot of times the scope of the game increases to match them erasing any games that you might have that mean that you might get from having the tools and we were already a bit in over our heads learning how to make our first xbox game so i felt like we should just push forward for splunky 2 though we had a bigger team and bigger plans and i think we're a bit better prepared so good tools made more sense to me and the communication was also going to be more asynchronous and more complex given

um the time zone differences and and uh the complexity of the game so blind ended up building among other things ways to set all the different variables in the game you know you could create whatever you wanted at whatever time to test different scenarios you could record replays slow the game down zoom in and out pan the camera etc aside from allowing us to test all these all the different scenarios that might come up these tools also made it much easier to grab screenshots and take footage from marketing which definitely can't be discounted and takes a lot of time for level design we didn't actually have an in-game editor but blit did externalize the level so that i could edit them in a text editor and use all kinds of flags to make my life easier and at this point i'm not actually sure if a real editor would even be faster for me because i'm i'm so used to doing things that way

from there we more or less executed on the design that i had in mind from the beginning i used google docs to write a design document for the game and that expanded to be a kind of road map for the development of uh as time went on we primarily use slack and the comment feature in docs to communicate with occasional video chats to talk through particularly difficult problems in any game project there's going to be some amount of exploration of ideas but uh and you know spelunky 2 is no different in that regard but i'd say that this was mostly a project of execution compared to spelunky classic or even splunky one which isn't to say that we didn't have anything to figure out there was plenty of that so much that it'd be impossible for me to cover it all in uh this talk but the unknowns were relegated more to the details i would say than the core ideas we had the previous splunky games to

look at and the places where we wanted to make improvements were relatively obvious com compared to the previous two games the biggest challenges i'd say that we faced during development were these all of these were things that we worked on throughout the entire development and only came together toward the the very end so first of all the art style i had a pretty good general idea of what direction i wanted to go with the art style spelunky had a look that was both cartoony and detailed which matched the arcade meets roguelike mechanics really well but uh the detail you know i added the detail without a real plan and i also had to repurpose a lot of the environmental art because of lack of time so as a result every area has this kind of rocky crunchy granola look to it you know which i think is pretty

cool actually but for two i wanted the detail to pop more in some parts and recede in other parts i wanted there to be i wanted to be more of a conscious decision and i also wanted each area of the game to have its own visual personality with distinct color schemes and patterns i wanted each area to really stand out on its own and i was inspired a lot by retro platformers that have the like wallpapered floor um you know as far as as far as that design went but it was a lot of extra work and it took time to find a good balance between easy to read and interesting to look at i started with something more on the cartoony side of the spectrum but over time i pulled it back to something that was closer to spelunky one and quite a few players felt like the graphics in our early game trailers looked flat and compared to the final game it definitely was

but you know compression of video especially if you're viewing it like in a small window or something like that also destroyed a lot of the detail uh you know with trailers although you always want to give off your best impression it's also just a really useful time because uh to get feedback because it can be some of the only like honest feedback you get from a wider more general audience um until until release of course design wise i probably the hardest time with quests i was really excited by the prospect of quests because it was an opportunity to add interesting new characters to the game and weave some static stories into the emerging ones the tricky part though is coming up with enough variety in the quests in a fast-paced arcade game that doesn't have a lot of text in in rpg you can kind of get away with

like a dozen similar fetch quests or kill the monster quests because you can add interest through descriptions and dialogue and things like that um which we didn't have access to in a game like spelunky ultimately i tried to tie each quest to a different system in the game like mounts or shops or gambling getting in trouble things like that but it took quite a bit of work to figure out exactly how each quest would play out and in what levels also given spelunky's design it wouldn't have felt right if the quests were just shoved in your face they had to be optional and feel well integrated into each level and that also took time as well and actually it's funny i was taking a screenshot the screenshot for this slide i wanted to get a screenshot of yang just giving me the quest of finding turkeys but i ended up accidentally

knocking a caveman onto yang and making him angry at me and i decided to just roll with it because it's like totally such a spelunky thing to happen regarding liquids i really wanted to make them more dynamic in spelunky ii in splunky one liquid could move around but in a very stilted way that still relied on the grid in general the grid the tile grid is an important part of spelunky's movement and level design that makes it possible both for me as a designer and for the player who's playing to know exactly where you can get to just by counting tiles grids are a very video gaming convention which actually i like but you know when there's a nice opportunity to break it up i like to take it spelunky for me is a game that is always sort of straddling two different concepts like that at once arcade and roguelike cartoony realistic grid-based and dynamic

from an art uh both an artistic and technical perspective liquids were really tricky to get right for one thing they're hard on the cpu early on we decided to limit them to the front layer of the game for that reason you can't have liquids in the in the back layer and throughout development we played around with their like the size and shape of each liquid blob to find a balance between realism and performance and we especially had to you know think about online multiplayer in that regard so it took some experimentation to get the right balance there and also to make make sure that the look match the rest of the game and it also took time to just figure out like what we could do with liquids with with this this type of liquid physics in in spelunky and in the end i think we came up with some cool ideas like fountains and and the lava

mander who can spray lava on you but you know there's probably even more potential there then you know there's online multiplayer which is always hard especially for a small team i know guillermo from bloodworks is giving a talk about the netcode specifically uh as part of gdc so if you're curious about the details of that please watch his talk from my understanding they evaluated the game and felt like a rollback method would best serve our needs given the size of our team and the game's precise platforming nature among other things but to make a long story short blit had to balance implementing my request for more complexity in the levels and also implementing the game or i'm sorry also optimizing the game to make sure that online multiplayer would work in the end although it may not seem like a splunky 2 is a really intensive game much more so than a lot of online games

like at its very worst there can be four players running around and lots of objects moving and colliding in the entire level across two layers with liquid physics and yeah the uh uh a big problem is is also just that right like all these things happen have to be able to happen at once and be updated every frame across the entire level so we we can't just ignore what's off screen for performance reasons unfortunately even with all this planning and hard work the online was uh pretty rough on release so it really goes to show you how tough a problem it is and uh you know i'll talk i'll talk more about release um when i get there the final developmental challenge was really just the sheer amount of content there is in the game much more

than spelunky one i mean just the player characters alone each one has over 100 frames of animation including extra animations for things like swimming and holding items that didn't exist in splwonky1 so about a year before release i i hired justin chan to help with illustrations which was definitely a big step for me because i was used to doing that artwork myself in all of my previous games but he did an awesome job and i'm honestly not sure how the game would have gotten done otherwise so you know a lot of credit to him and also credit to to my wife francis for for pushing me to get help with art in the first place okay so it's talking about spelunky 2 but i need to briefly mention ufo 50 because for the first part of spelunky 2's development i was working on ufo 50 at

the same time which now sounds kind of crazy even to me but it ended up working out pretty well i think because the two projects complemented each other um really nicely and and scratched different creative itches and uh that picture that picture of anna right there on the left is is by justin so here's a comparison of splunky 2 and and ufo50 splunky 2 was a sequel it was an existing idea we were expanding on the artwork was hand drawn and very intensive on the other hand ufo 50 is 50 completely new retro games with pixel graphics that i find you know very breezy and fun to work on compared to splunky 2's hand-drawn art of course the time spent on one project definitely limited how much time i could spend on the other but the benefit which was huge was that i always felt very refreshed on both projects

when i was working i was uh i was much more focused than i than i usually was working on like just one big project at a time from my experience you know it can just be really easy to get lost in a single project if that's all you're thinking about all the time so you know even if you're not working on two projects at once which i don't necessarily recommend at the very least it's important to take breaks and reset your brain once in a while for the most part i treated ufo 50 like a serious hobbyist project whereas splunky 2 was more like my day job and at one point i literally worked on spunky 2 during the day and ufo 50 at night and even used different computers to to work on them to help keep them kind of separated in my mind but there were months where i had to focus on one project over the other because of a deadline or a particularly challenging problem uh eric was also working on both projects

and i'm guessing he had a similar experience as me um other than me and eric ufo 50 had a different team than spelunky 2. but we had a really fun time at pax one year when both games were being shown and and both teams got to hang out together in terms of design as i mentioned before i wanted to preserve the unique qualities of spelunky one compared to other roguelikes which are also qualities that i just personally enjoy in games period to sum up it's a feeling of entering a world that feels like it's not built for you but offers you a lot of agency as the player to explore it it's happy for you when you make progress but progress isn't the main reason for for it existing um there are consequences to your actions in these worlds and that makes you really care about what you're doing

uh i think that's really important and you know those are the types of games where where i feel the most immersed irrespective of how good the graphics are or you know what the style is and i've been thinking about an easy way to talk about the style of game and i uh that i really like and i came up with this word spikey like this spiky sea urchin that i think works really well for me so i think spiky games are often thought of as punishing but to me the difficulty while it's an important part of the design ethos is in service of the goal rather than the goal itself the real goal i think is to put the player in a state of focus about the game and to really care about what they're doing at any given moment um it's to make them feel like the world

existed before they arrived and that it will exist after they leave to make it feel real uh winning or finishing the game in my opinion is not the main goal of a spiky game even if you know that's ultimately what you're working toward the opposite of spiky design could be called soft design and you know just to be clear i think that every game has some soft elements to it and some spiky elements to it and there are definitely times where i enjoy spiky design and i feel like playing a i'm sorry where i enjoy softer design and i i i'm looking forward to playing a soft game for spelunky 2 i actually wanted to increase the softness compared to spelunky one i just wanted to make sure that i did it in a way that it didn't make the game less spiky

so for example you know we we expanded the base camp and made that a place where you could relax practice your moves and feel a sense of progress as you unlock more of it at this point you might be reading these descriptions and thinking well you know aren't you just talking about casual and hardcore games well even if there's a significant overlap between soft spiky and casual hardcore i feel like the latter terms are just too loaded and they're too focused on difficulty and certain genres now difficulty is maybe the easiest way that we have right now to achieve the feeling i'm talking about but i wouldn't say it's totally necessary to to spikey design and actually you know i i get a lot of spiky feelings from playing animal crossing new horizons where i feel like some of the user

interface designs are purposefully a a little clunky just to try and force the player to or you know encourage the player let's say to take it slow and treat the island like a place to live rather than a place to min max and optimize and you know get the biggest house as fast as you can so like you know one example is the fact that you can't eat a hundred pairs at once right i feel like they may put that in there just so you can more easily enjoy that island life and and uh and relax unsurprisingly not everybody likes spiky games it's like bitter or spicy food it's an acquired taste that you have to build up people who enjoy spicy food really love it though food that's not spicy can even taste a little bland to them

it can be hard for people who don't like spicy food to understand why spicy food fans love it so much if you're just starting out on your spicy food journey you can't taste the flavor just the heat and the pain similarly spiky games generate a lot of enthusiasm from their fans but for people who aren't there yet they can just seem hard and it doesn't help that just like with spicy food there are people who are just there for the challenge and bragging rights and there are peppers hot peppers and games that are made to cater to those types of people not that there's anything wrong with wanting challenge for the sake of challenge either but it does make things that much more confusing when people are trying to evaluate spiky games again for me the difficulty is only one part of the equation it's the heat part

of spicy food i don't eat spicy food to feel pain but the pain wakes me up and it's the gateway to interesting flavors that you can't find anywhere else the flavor is what makes spicy food good and it gets easier and easier to withstand the heat the more you experience it and the great thing is that there isn't just one spicy dish that you can eat to improve your spice eating ability there are lots of delicious options eat one and you'll make it that much easier to appreciate another same with spiky games uh learning to enjoy them will open up an entire smorgasbord of great games to enjoy and if you like these kinds of games and want to make one it's important to think um or sorry it's important to understand their flavor and think about what kind of unique

flavor that you want to bring to the world and here's an image showing spiky and spicy together which is making me really hungry and also want to play some games so all right back to spelunky 2 specifically we've got all these things we've built up across the previous two games for me what i wanted to do with the sequel was just to take that and make it even bigger and feel even more alive what are things that we introduced in spelunky one that are cool and can be taken even further uh in spelunky too so one thing you can do for example is just make things bigger in mario they started with bullet bill and later added bonsai bill

in doom 2 they added the super shotgun which has two barrels well you know we can do a similar thing with splunky by taking the small crush trap and making a bigger version of that and since we had a chain in spelunky one where you collect special items to get to hell let's make that secret route even longer and crazier and weave it into a branching path system some of the joy of sequels for me is just seeing things taken to their logically extreme conclusion we have that freedom you know to do that because no matter what the original games will always exist on change and that gives us the opportunity to just explore the possibility space however we want there are a lot of other little things that i could talk about like having the cavemen pick up items and

talk to each other for example was added pretty late in development and i think that for how much work that it evolved it added so much personality to the game and does so much to make the world feel alive and interactive but it's the kind of thing that's hard to think about until you have enough of the game's foundation built i will talk a little bit about the level generation though since we did make some interesting changes to the basic algorithm to accommodate new features like the back layer and also to make levels feel a little more natural the way spelunky ones level generation worked is that we take a typical level of four by four rooms and we generate a path from the top to the bottom marking each room as a certain type like entrance exit normal drop side room etc

then we generate all the rooms based on what they got marked as of course it's more complicated than that especially since there are special rooms like shops and altars but that's the gist of it the biggest change to the level generation was the addition of the back layer in spelunky 2. this happens after the front layer path is generated and the rooms are turned into strings but before they're actually turned into tiles and the reason we need to do it after we generate the path and the strings is that we need to know the exact structure of the front layer so that we can connect it to the back layer through doors then after generating all of the back layer rooms we connect them with special passageway rooms and shops and treasure caches can also be generated there as well

there are also uh back layer rooms that always appear in specific levels and specific configurations like madame tusk's palace of pleasure and so we added special dual rooms where i could define the front of the room and the back of the room together and this is just a a drawing that i made to send a blit to show them what madame tusk's palace of pleasure is supposed to look like the other major addition to the level generation are machines which are inspired by the machines from the classic roguelike brogue uh so you know they're called machines but they're not like actual machines like robots that you encounter in in the game embroid machines are just a special generation

a special part of the generation that is used to add interest to dungeons um after the basic room generation and they might include special treasure rooms or swamps or chasms with bridges across them stuff like that some of them are just for flavor in spelunky i realized that once we had the path set we could create rooms that were extra wide or extra tall and these machines were used in conjunction with the back layer to really open up what was possible with spelunky's level generation so they weren't used in exactly the same way as as brogue but it was the fact that brogue added this kind of extra step in its level generation did did inspire me and brogue by the way is a very fun game and i think it has a great balance of simplicity and of playing and complexity of learning and i

still think that there's a lot that we can learn from traditional roguelikes all right so now it's time to release the game toward the end of 2019 i spoke with eric and it felt like we need to make a decision to focus on either spelunky 2 or ufo50 and finish one of them partly for our own mental health as the fun of working on two projects at the same time was starting to wear off and the pressure to release was really starting to build delays are common of course but having two delays going on at once wasn't feeling so great and so we had to uh we had to make a decision and ultimately splunky 2 seemed like the game that was most likely to be finished in 2020 the road map for finishing it seemed clearer and it felt like if we buckled up and focused on it for a year we could get it out it also had the most people waiting for

it being a sequel and a known property and all that so that was that the rest of the ufo 50 continued to work on ufo 50 while eric and i moved over to spock e2 full time to finish it since we were already working remotely it didn't feel like the pandemic had a huge effect on our work in the last stretch of development um in fact you know 2020 ended up being a very packed year for games with lots of big games indie and aaa coming out last year the ps5 and the xbox series x also dropped in uh in 2020 so for teams that could work while remotely it may have seemed like a good time to release since so many people were stuck at home looking for something to play you know and we were second home looking for work to do also so uh

our but for us you know i think it was mostly a coincidence that we released in 2020 we had wanted to release splunky 2 in 2019 and when that deadline passed we were just determined to get the game out the year after but uh you know aside from being just a big year for games in general it was also a big year for roguelikes right aside from spelunky 2 there was hades risk of rain 2 rogue legacy 2 noeda atomic crops and probably more and they all uh a lot of the big ones always ended up coming out pretty close to each other um in the in the summer and fall so you know was i worried about that honestly yeah it did make me a bit nervous i was also concerned about cyberpunk which was uh originally slated for a september release but you know i think all releases are like that you can spend years working on

a game and then when it's time for the game to come out there's only so much you can do uh to control what's going on around it and uh you know ultimately for us i don't think it would have made sense for us to delay the release into 2021 just to avoid other roguelikes and there's never any guarantee that waiting is is going to make it any less packed um you know i do get asked about early access every now and then because it does seem like a popular and successful way to make big roguelike games but you know i don't know for me it just didn't feel right to release spelunky that way i i think there is a difference between updating a game in early access and updating after release and spelunky as this spikey arcade roguelike where i had a really strong idea for what i

wanted it to be it felt like it needed to come out and and be fully released before we got serious feedback from the public um i wouldn't be averse to trying early access for games in the future but it would probably be a very different type of game something much more open-ended like a like a simulation or something like that so you know ultimately it felt like we had to put our faith in both the game in spelunky 2 and in the roguelike genre as being something big enough to have all these games come out at the same time and do well so we we set our release release dates and went for it uh you know by the time september rolled around we we ended up having a staggered release on ps4 and uh and then a couple weeks later on on steam

and we focused on those two platforms because you know we had we had good relationships there and the majority of our audience of spelunky fans were were playing the game there you know the prevailing theory of course is that you release simultaneously on every platform at once and you know i'm sure that's great but given the size of our team it just it just wasn't possible especially since splunky 2 had online multiplayer for me personally this was my first game release in seven years so you know a lot had changed i'm i'm like an old man dad now and uh you know the scene has changed a lot too people love roguelikes now and they know what spelunky is press actively sought us out for for interviews and and reviews which is cool uh social media is

much bigger now so you know it gives developers and players and then media just a much more direct con connection indie games are more serious now too with like serious studios and publishers in the in the indus indie sphere doing some big things so you know in short we were where we wanted to be and i think we felt very fortunate but also at the same time the expectations were really high and i think it was it was overwhelming how many people we were potentially connected to now and and uh you know for me i realized that this was this was another leveling up moment starting from when when i was just working on on click and play games as a kid and uh you know we had to step up to make sure we were taking advantage of of these opportunities that we had that we were lucky enough to have

so there were definitely some hiccups with the release as you can see here aside from the usual chaos uh bugs and stuff like that the the online multiplayer launch on ps4 was much more rough than we expected when it worked for people it worked quite well but for many people it was pretty much unplayable or after playing for a few levels it would become unplayable so this forced us to delay the online multiplayer release on pc while blood worked on fixes and then added more servers so again guillermo goes into this in detail in his talk and it's really actually quite interesting so again check that out if you want to hear more about it suffice to say though video games are very complex and adding online multiplayer to a game is just another

layer of complexity you know and we tried to be very clear that we were doing our best to fix it and just be very direct and upfront and thankfully you know most players were were understanding once the game uh released on steam it was easier to to get feedback for blit and narrow down the various problems and you know quickly test possible solutions so big thanks to all the spelunky players who helped out during this time with their help blit was able to make the necessary improvements to online and yeah thankfully it started to work well on ps4 and we're also able to launch it on on steam by christmas time and all in all i'm i'm i'm happy with how things went in hindsight you always realize just how ambitious you were

trying to make a game and we really tried to do a lot with spelunky too uh and you know in the end i don't feel like we had to give up on any of our major plans the game came out how how we wanted to how we wanted it to and you know more or less how i envisioned it uh a dense spiky rogue-like platformer and you know the response from players has been has been great um and you know i'm very happy about that i think with any sequel they're going to be players that are going to be disappointed um but for the most part it seemed like the game was being enjoyed by veterans of the series and new players alike compared to spelunky one uh spelunky two's lifetime revenue so far is actually getting close to overtaking spelunky one after just less than a year which is pretty incredible

and not only that but you know the release of splunky 2 actually uh i think created buzz for spelunky one and got more people excited about spelunky one also so you know that's it's not bad at all for for a game that that i honestly think is weird um in some ways it straddles a lot of different concepts that i think can feel can feel kind of opposed at times as you can see here and you know it was a really big year for roguelikes and people naturally wanted to compare them but uh you know if you're in the same conversation as a game like hades i think you're already doing something something right it's a it's a cool game but spelunky it's in my mind it's a very different game from hades which is a very different game from noita which is such a different game from risk of rain

etc so by any other metric it really seems to me like all of the major roguelikes that came out in 2020 were quite different from one one another and they all found sizable audiences that you know really show how much room there there is in this genre uh some genres i think really lack the flexibility that roguelikes have and they become saturated very quickly but that that's obviously not the case here in my opinion so yeah i think that given our experience in what i've seen from other games indie games sequels make a lot of sense if the original game was successful and you're not burnt out on the concept in the past indie sequels weren't really a thing partly because every indie game developer back then was just we're figuring things out as we went along there weren't many established indie studios and

i don't think anyone was really thinking like super long term about their career we were just you know trying to get the game that was in front of us done by the end of a big project it was pretty likely you'd be burnt out on the idea and want to shift gears completely once you're finished uh you know and i think indie developers in particular value novelty but players for the most part are pretty excited to get more of what they already love i i definitely had that itch to work on new ideas while i was working on spelunky too having ufo 50 to work on obviously helped a ton but the crazy thing is that it didn't make that feeling go away completely so it really goes to show you how strong that feeling of wanting to make something new is no matter what you're doing so you know what i do is if i have a new idea that doesn't fit what i'm already working on i write it down no matter how silly the idea is

and sometimes that's really all you need to get it out of your system and you know personally i really enjoyed being able to connect all three spelunky games together and make a series at times it felt like taking an old childhood drawing and getting to redo it as an as an adult which was pretty cool and you know looking back at it now like yeah it just splunky just feels feels really big to me so while sequels make a lot of sense from a commercial perspective i think they can be really fun from a creative perspective as well so you know do i necessarily recommend doing all the things i described in this talk like making a sequel or not changing the basic you know uh formula of the sequel too much working

on another game or a collection of games at the same time well you know it worked out for us but splunky 2 was a unique project ufo 50 is a is a unique project all games are unique i think they require uh they required unique circumstances to succeed basically and i i think it would be hard to replicate what we did exactly every game developer and game project is is just a little different uh however i i do think that there are some things that that may be helpful to other developers just some general things that i i thought about and learned as i was working on splunky too the first is that you know making commercial art it's like it's like a dance between you and and the audience where you're trying to find where your interests overlap with theirs and you're also developing your skills at the same time

so you know we're talking about a an activity that's incredibly complex and long term so i think it's important to realize that splunky has now spanned three games across a decade of work when i was working on splunky classic i had no idea if anyone would like it as much as i did and i'm glad at that point that i only spent you know six months to a year working on it before releasing it as freeware so if if people hadn't liked it it would have been back to the drawing board and the longer i spent on it the more opportunities i would have lost with swanky 2 i feel like i was able to realize the potential of the concept more fully but in a much safer environment where i knew people liked the concept so you know you can take multiple games to experiment and refine your craft i think that's something to remember two it's kind of related but these days i focus as much on the process

of making games as the games themselves because i i have the experience now to know how hard game making is at any level i don't just make a game because i want to make the game i make it because it's also the right time to make it and the right people are around it to help me make it and i never assume that a game is just gonna get made out of sheer will a lot of the decisions that you make in the conceptual phase will either help you or haunt you once the development starts um so you know i think it's i think it's important to uh to think about the process three there's there's not one way to do this uh like i said i've you know i feel like i was in a unique uh position i am in a unique position where i've been at this so long and i have the opportunity to work with so many cool collaborators but i still feel like

very indie in my development um you know i'm still setting up my own websites and store pages and doing art and design and even some programming whereas a lot of indies are much more specialized and working in studios where they work alongside the same people for many many projects and then you know there are game developers who are just totally solo so indie game development to me is is as much a lifestyle as a job so find a way of doing it that works for you and also realize that figuring out how to do it is a journey in and of itself and then finally yeah i mean roguelikes i i really believe will continue to be a strong genre for a long time if that wasn't obvious before then i think 2020s in the roguelike bumper crop should

definitely make that clear if you want to make one you should make one but uh you know that doesn't necessarily mean it's it's going to be a success it it just means that there is room there is room for more and then finally you know i did promise i would i would talk about the possibility uh of uh more spelunky's since you know we're talking about sequels and all that and honestly you're probably expecting this but yeah i don't have any idea right now right now i've got my hands full with more spelunky to work and also finishing ufo 50. after that i think i'm going to want to go back to solo development for a while and experiment again with new ideas and uh you know maybe new new game engines things like that so definitely no spelunky 3 anytime soon but never say never

and thank you so much for listening and i hope you found this talk to be interesting and helpful you can always reach me in these places and yeah have a great rest of your day thanks