Cuphead's Animation Process and Philosophy

[Music] welcome thanks for coming my name is Jake Jake Clark and I'm one of the animators designers concept artists and idea people on the upcoming indie game cuphead which is still not out yet apparently games take a long time to make nobody told us so sorry but you'll find out very soon why it's been taken so so damn long so a little bit a little bit about me I went to school at a Capilano University in Vancouver I was trained in traditional paper animation and then oh if anybody's interested in learning to the animation Capilano University in Vancouver is amazing I definitely wouldn't be here without it so after graduating I worked on a few 2d cartoons and Vancouver's TV animation industry mother up for Hulu turbo FAS T for Netflix breadwinners for Nickelodeon and then out of nowhere one day I got hired to work on cuphead and I've been

doing that for two years now or a little bit over two years and it's actually my first job in videogames and I feel very lucky so what does cuphead it's a 2d action action game the gameplays inspired by games like gun star heroes contra hardcore Metal Slug just games from the 16-bit era and it styled to look like a 1930s cartoon like Betty Boop Popeye old Mickey Mouse cartoons that kind of thing it's the love child of Chad and Hurd moldenhauer who were my bosses and they grew up playing those those kind of games and watching those kind of cartoons and that's why the game looks and plays like it does we animate it and ink it on paper traditionally and the backgrounds are traditional watercolour paintings the the music is like live recorded by a big jazz band so we're really going all out to make it as

authentic to how they would have done it how they would have made a game in the 30s as we possibly can without going overboard so I'll show you the trailer we released in 2015 to give you you a much better idea of what what this game is about [Music]

don't do the bad thank you very much so this is the murmur that you just saw and I'll take you through her being just an idea to a finished design that's ready to animate so first of all is just what makes a 1930s design so these are what I always keep in mind when designing I just kind of they're just broad general rules that I try to follow whenever I'm doing any kind of design for the game of course any rule can be broken but for the most part 30 designs usually air toward simplicity they don't tend to make overly detailed designs they're usually made of solid shapes and

why I think that's really important is that like any 30s design can be turned in space perfectly without needing to do any kind of cheats so you're almost like trying to draw a 3d model in a way so it's not like a lot of modern cartoons like because anybody here seeing clone high which is a fantastic show but it has a lot of like stylized corners and edges on characters and things that like you wouldn't be able to turn Abe from Clone High like 20 degrees to the left if you know what I mean he just has a lot of stylistic tricks and they don't really translate that well to three dimensional turns a lot of most 3d characters are usually pretty rounded a lot of round shapes round lines sometimes sharp fingers sometimes sharp air or sharp hat but for the most part a lot of round shapes and then the term I'm sure

everybody's heard of is rubber hose which is sort of like most of the characters they're almost it's almost like they're made of rubber both in that you can you can really stretch them a lot and squash them and play with them that way when you're animating but also a lot of them have kind of bendy Tube limbs and the lastly limited features by that means by that I mean like we're trying to make this game look like a 30s cartoon and how do we do that we do it by referencing all the different shapes that they used to use in the thirties the different way they drew hands the different way they drew eyes so in a way when you're designing for a 32 cartoon it's almost like frankensteining together design in a sense not entirely but most characters have a certain way of doing things so the first thing I do before starting a design is just think things think um yeah I think about what would contrast the other bosses because

we want all the different bosses in the game to be very unique to each other and what has been done already both in our game and in other games because we don't want to do anything that's been done before and then what would excite me is the players so I actually imagine myself playing the game I try to see in my head the little cup head guy running around shooting and try to imagine what it would be exciting to actually interact with and then I think what would I enjoy animating because of course you want your design to be as good as possible but you don't want to put any ridiculous amounts of detail that make it very hard to to turn in space into animate for example I'm showing you how to design the mermaid one of my early concepts or when my early ideas was to have her have a pirate ship on her head and if you imagine trying to animate that and turn that in space it would not be a great time so you want to be animating something that you enjoy your animation is going to be better not going to be as much of a headache like if you're trying

to do 3d animation with a rig that you hate your animation is just not going to be as good as if you're using a rig that you like right and then I think what would have lots of gameplay an animation potential so I just want the character to be capable of a broad range of gestures with the body and actual subject matter of like what the attacks would be so these are the very first designs that I did so I was going for strong fierce wild magic magical I wanted to have octopus hair and look as if she ate pirate ships so I first did these before doing any reference which like reference is very important but I do think there's kind of a value to taking a swing at something like a design or an animation before you reference just just because prior to referencing you have an idea in your head of what this thing is that's based

on everything that you've seen so for this the idea in my head is based on all the mermaids I've seen all the strong characters I've seen both in cartoons games and real life all the fierce things I've seen all the wild and magical things I've seen and so on so everyone in this room if tasked with designing a mermaid everybody has a unique idea right after that and it's very rare that you would you would do just hammer out that design without doing any reference and have that be the final design but something something valuable to to starting without reference so I got some notes back to try her to look a little bit more like Betty Boop or a little bit more like olive oil so I looked at what makes Betty Boop Betty Boop I looked at her the shape of her eyes and she had tiny nose and tiny mouth and I thought quite a bit of that could apply to the mermaid design I didn't really look at olive oil

and just didn't really think that she fit that well story Chad and then I opened up reference to other game or other other characters that I thought could help like Eliza from fantastic journey which is where I got the pirate ship on the head idea splash from one from Mega Man 9 and then oh I was also told to try some different things on her head other than octopus hair and I just felt Wonder Woman sort of fitted the vibe that I was going for then I did some of these where I was experimenting with what she's wearing on her head a lot of these have those Betty Boop kind of eyes and quite a bit quite a few of her features she's definitely the primary reference for the character yeah most of these designs are just heads which in hindsight I probably would have done more full body designs but this was two years ago and I'm a growing boy still learning so then Chad

who is the art director on cap head he took a couple of the drawings and he played just played around with maybe she could have skinnier arms maybe she could have a different different hand and then on the bottom left is just a drawing of his where he was experimenting with some of his ideas then he gave me there's a girl from the moth in the flame on the top right we thought she has a cool kind of eyes that could could fit the character and then I was mentioning I wanted her to be a bit sicker so they gave me an early Betty Boop design where she where she's with their co2 to reference proportions so yeah anything that they can possibly reference anything we can reference we just do it there's no reason not to we of course don't steal stuff but you got a reference if you want to make it look like a specific style of course so they went with octopus hair which I was very happy about and then yeah I did just a

couple of full bodies and a couple of heads I find that the longer the design process goes on the fewer drawings you have to do because you you start out trying to figure out what it's going to be like what's the subject matter who what kind of mermaid what does she have on her head and so on and then you kind of get to just specifically just how big or different things how bigger her eyes and things like that you kind of narrow it down and you just and I find I get a better sense of what they want as time goes on and so they chose the one on the left here and Chad did a different chest piece it's a little bit more like Ariel from The Little Mermaid and then this is actually the final design we don't I can't actually speak for the other animators but before animating I don't make we don't make like a full turnaround model sheet kind of thing we tend to kind of treat the idol ads

the model sheet because everything goes into the idol everything comes out of the idol and generally in the 30s those cartoons weren't super strict about models like like a Disney Princess cartoon would be very dear so I'll show you the process in doing her idol animation so the first thing is I just like to figure it out before I start drawing as much as I can anyway because we're animating on paper the stakes are pretty high like if you're 10 hours into an animation then you find out that you made a critical error it's going to be a really really long night so you want to know that it's going to work as early as you can and it's just way harder to change things if you're doing it on paper so I just visualize I just try to imagine try to see in my head what what would work what and then as what has been done before and so on and then then

I acted out I don't really think of it as acting it's more like like a little kid running around in the sandbox pretending they're a monster kind of thing and then you just yeah I wouldn't call that acting but you are it's just a really good way to get a starting point you just try things because a lot of what you're going to be animating is going to have a head it's going to have arms it's going to have legs even if it doesn't you can still use your body you can kind of use your body as a tool just to test different animations on the fly just to get a starting point and then points to hit so what I mean by that is that I'm I try to I try to figure out what is non-negotiable what is definitely what needs to be in this animation for example for the mermaid she's swimming in the water the background is panning so she's always moving she's swimming and so that means

for sure she has to do like a hip a hip swing she asked you so instead of one like worrying about what she's gonna do with her lower lower lower half of her upper body you know it's more just how is she gonna sway her hips because I already know that she's definitely going to sway her hips and then I also in other points that I needed to make sure to hit is that she needs her hands to be at the ready because if you're fighting a tiny plane it's flying around you need your hands ready to do things you don't see a boxer go into a fight with their hands line their back and then I wanted it to have a dance ish kind of quality because a lot of thirties cartoons a lot of the characters are sort of dancing and bobbing most of the time really so a lot of the animations in Cup had have a sort of dance like quality and then I thumbnail and reference generally but I did not for this one I felt like I had a good enough idea and I just kind of

started going straight into the rough but usually I would some male and reference those are very good ideas so this is the the first drawings that I did for the idol I leave it pretty damn rough or these are done in blue pencil can't really tell of course but I leave it super rough just because I don't want to be a huge time commitment so in case I have to do it again it's not that big of that big of a deal but I just want to clearly communicate exactly what's going on um and then I took those same four drawings and I tied them down with the file into the final model was the final design like I added little little nubs on the tentacles at the top and scales on the bottom and all that but I want to have like a solid skeleton to work off work off of so I want to just lock it down and a few drawings keep in mind how all the different elements are going to move and it's again my kind of the hardest part of animation for me is the entire time

before I know that it's going to work so I just want to lock it down make sure that's going to work and then the rest is just I it's a lot less stressful that way so then I did four breakdown I'm not here to tell you what breakdowns are you know I emphasize the drag of the tentacles and offset limbs and all that stuff and then I did eight more drawings to put it on twos this is a sixteen drawing total and each I did putting it on two twos every single in between was basically just another mini breakdown but then once it's on toons taking it to one if there aren't a lot of exaggerated arcs then I basically just turn my brain off probably put on a podcast or something and just straight in or halfway in between between everything I don't have to keep in mind exactly how everything is turning in space and things like that it's just if the lines

here and then the lines there the in-between lines going to be right in the middle that's usually what they did in the thirties anyway we just definitely don't have time to be doing even more many breakdowns to get it down to two ones and then here's the final animation it's 32 drawings on 32 pieces of paper which would then I would then scan them I would send them off and then they would be inked by hand onto 32 more pieces of paper and that is why we're not done the game yet because we're ridiculous but hey it's coming out soon now show you someone from the tombstone box who is the final phase of the slime boss he's a little slime and then a big slime and then the big slime is crushed by his own tombstone and you fight the tombstone it makes sense and I'll show you his slam attack so this is what I got from Chad

they just just a quick thumbnail to show exactly what's what's going on and then I didn't have to do a whole lot of planning and thinking about this one I mean I'm working with just a slab of stone this says you make it go down it goes whack that's you can't really read that in any other way don't need to do a lot of thumbnailing or anything so I did just the main extreme for at this point they would test to see how it's going to work in the game how it's going to visually be laid out how it's going to affect the gameplay how many frames I'm going to have to have him slammed down to the final pose which is where it would actually do damage to the player so we figured out all that kind of stuff then we lock it down and there in the rough stage and then I go ahead with it but this is the this is the rough it's I think it's just for drawing maybe five again it's just to show exactly what's going on I added a screen shake in TV paint which is the program that we use to shoot or that I used to shoot I think

a couple animators do something else but yeah just to help emphasize what's going to happen because it's actually going to shake the screen in-game as you'll see later so then I added all the details locked down and here it is on twos and then here it is on one so an animation like this none of the none of the in-betweens would really be mindless it doesn't really work when something is moving in this kind of way and here it is actually in the game this quick little thing I want to talk about with the antics how he he leans back there before he slams down I wanted to specifically make sure that he leaned back enough that I would get enough perspective change that you would see the dark color under the base of it you would get you see the the white rim line underneath the top of the thing and the light spot underneath where the slime protrudes because I'm the point of an antic generally and something like this

is to get the players attention you want them to know that they pay attention they need to get the hell out of the way or whatever it is you just want to get the players attention so I mean I could have just had him squashed down or stretch up a little bit for the antic but if I do it this way then I not only get change of shape but I get change of color and change of perspective as well because I find kind of just the more something changes the more visual gravity it's going to have because you want the player to notice as much as you can help okay here's these two the garbha and it's actually a world first reveal nobody has seen this guy yet you're very very lucky people [Music] and I'll show you his death animation so yeah I just started out I was given quite a bit of freedom with this one which is really nice when that happens so basically I sort of just went through the same design the same process I go

through when designing something so I look at all the other death animations in the game all other death animations and other games and so on and just see what's been done what would be interesting what would get me excited and something like that so the first idea I had was the ashley part of his head lights up and he sort of gets smoked from the top down and he's left with pile of ashes but it's worked better for a different animation that maybe he could be leaned back crying ashes pouring out of his head maybe hunched over forward little bits falling off of his his head maybe burst into flames and scream smoke away the pain of losing cough up cigarette or split in half and poor ash all over the place but it's way too gnarly and the game is rated E for Everyone so we're not doing that one so this is the one that we went with yeah definitely a Monty Python inspired idea and we went with it because one there is

no other death in the game death animation in the game like it but also it gives kind of a cool sense of scale because you're tiny compared to this giant cigar that you're fighting and then the cigar is tiny compared to whatever's crushing the cigar so you feel extra small so I was given just rough thumbnails to test it with the background see that it would actually see how it would work on the background platform and then Joseph Coleman is the animator of the secret character that is stomping him so he sent some leg designs and we picked the middle one then I did just rough outlines basically of the extremes of the thing so that we know because the cigars on a platform and we need to meet we need to know that it's going to match the background properly and I did just the main keys oh I did some reference and discovered a

fetish of crushing crushing things like cigars which was quite enlightening you're welcome look it up very interesting anyways so here's the the rough animation and with this one I felt pretty confident in it it seemed like there wasn't a whole lot of different ways that it could go so I didn't really bother like putting it on twos and then sending it off I just went straight for one that is that's what I did summer red wall and then so why do we do this to ourselves why do we animate it on paper why do we ink it on paper why do we do the background traditionally watercolor is why the lab jazz band we just want to make it as authentic to the 30s as we as we can where we draw the line is we don't are we color in Photoshop we don't ink on to transparent cell and paint on the back of the cell because that would just be completely insane just pure frustration and you wouldn't actually gain anything

from doing that but we think that you do gain something from doing it on paper I finally like for me anyways my drawings are just plain better if I'm animating on paper and the doing it with a live jazz band it sounds way better than if you try to get it to just sound like it was done in the 30s with MIDI instruments right and I also think you got a quality of actually doing real watercolor backgrounds rather than I mean you can make a digital watercoloring color looking background but yeah I do think you kind of lose something there and then yeah we want to help keep paper animation alive there aren't a whole lot of games or cartoons really these days that are made with it and also it's just the most fun way to do it like it's what the team wants to do it's we have a great time doing it this way more fun than we would if we did it digitally and we think that in this case it's worth it and like we really want to do it we really want to

make a game that's likely would have done in the 30s rather than just try to make it look like we did it whoops so live a 90-30 style does not plan mmm one second here there we go okay so yeah I mean the 30 style just perfectly fits a side-scrolling game a lot of the compositions in thirties cartoons they actually look a lot of the stuff takes place on a flat plane and it seems like I mean when I first saw the trailer for the game I thought why is nobody thought of this before like it seems like such a perfect fit but apparently they were the first to really go through with it and there's those create those cartoons are just super creative just one interesting idea after another and a lot of that kind of thing is not really in cartoons these days and then the what I think is the most important reason to do 3ds cartoons or the biggest benefit of

doing it in this style is that we have complete freedom over the gameplay the gameplay is not limited by the subject matter what that means is uh like anything can happen in the thirties cartoon you got the ghost man in the bottom middle he's pulling his head off turn it into a bottle pouring a drink porn it and neck dancing labs lamps all kinds of crazy stuff so we can just say like I want a bus that I want a bus that shoots something out of it or shoots three things out of his stomach in a spread and then those things explode and then three other things chase you around okay and then once the gameplays figure it out then we can fill in the subject matter after that so we could say it's a clown bear and a mouse opens up in his belly and bees fly out and the bees explode and then ghosts bees chase you

around that could happen in a 32b so we're just we could do whatever we want for the gameplay and we're not worth like if if the game was if it looked like king of the hill and it was in a king of the hill kind of world you wouldn't be able to do this kind of thing you'd be much more limited in what you can do you can make a sweet game they'll get me wrong but you couldn't have Hank he'll have a mouth open up in his belly and bees fly out it just wouldn't fly that's it thanks everybody follow me a tad Jake Clark dude on Twitter I'll be posting a lot of animations a lot of designs and concept art as I can I can currently only show what's come out in trailers what we've shown in gameplay videos and stuff like that but once I'm able to I'll be dumping basically everything that I like online but yeah that'll be pretty soon it's coming out this year we're getting real close and will will post about the release date when we know what's totally once it's totally nailed down we'll let

everybody know on social media and all that and that's it thank you very much and enjoy the rest of you camp [Applause]