Animation Bootcamp: Animation Prototyping for Games

[Music] so guys thanks for coming out I want to thank Tim and Mike for putting this together this is a lot of people this is really great so I know this was asked earlier you know how many animators we have how many how many game developer animators do we have in the room today and then students and then like non animators just here okay great it's a great turnout um so I'm Jay Haas belt I'm the lead animator at Epic Games and I want to talk about animation prototyping so they say that you teach best what you need to learn the most and I've been at Epic Games for 10 years and we never had a formal prototype process when I first started there the first game I worked on was unreal championship 2 and that game was heavily based on prior versions of unreal and a lot of gameplay systems were kind of set in stone there wasn't a big global game to prototype and as I was working on you

see - down the hall they were starting Gears of War and Gears of War if you don't know started off as actually unreal kind of gameplay mode and that evolved over time as well so even Gears of War never started out with this like official prototype phase of a game and then of course Gears of War 1 was successful we moved on to 2 & 3 and sequels don't really have a time for a prototype phase as well and recently we sold the Gears of War IP to Microsoft Studios and black tusk is working on it and suddenly we're going to work on a new game that's not based on a prior game and we had to approach this from a total blank slate and to prototype a game and this summer I had never done before and in such a capacity and before me working at epic I worked in games for 10 years and none of those games had a really formal prototype process and then that's what I want to cover today it's kind of

like my journey from discovering all that so to kind of give the essence of prototyping we played this game at epic and we kind of had this company meeting where we all came down to play this game and it's called the marshmallow challenge I don't know if you guys have heard of it but we had to break up in groups of four and the challenge was to build the tallest tower to keep the marshmallow on the top and we had 20 pieces of spaghetti some string some tape not much more and we had 18 minutes to do it so Celia who was hosting this meeting said go and so my group I think it was a program or a marketing guy and animator and as someone else and of course we all turned to the programmer because those are the smart ones right and he had a pretty good idea and we started getting ready to go the guy next to me kind of called out a flaw he thought he saw and that idea so he proposed his idea and

that looked good until the guy over here kind of called out a flaw and we listened to his idea got to me same thing I'm an animator I've worked on complex rigs I can design a marshmallow tower so I gave I gave my example well didn't we heard cilia say all right you got five minutes left we had been talking for 13 of those 18 minutes so it was like musical chairs we we were on my idea so we went with it we all it's kind of divided labor I was going to make the middle part just go over here was gonna make the supporting leg this guy was cutting the tape and we heard so you then say ten seconds left and we're just speeding along and we're looking around to see you the other towers and she counts down three two and we stick that marshmallow on top and we looked around we had the tallest tower it's great so as she's walking over our Tower just goes boom and it falls this is us working on it so we lost and these guys won and I don't know if you can tell from the picture but their tower is

about as tall as one strand of spaghetti I mean they spent 18 minutes they could have easily just like three strands in a marshmallow so we sat down and with cilia with these meetings things are never quite what they seem so she showed us this TED talk and it turns out this marshmallow challenge has been done all through the country different schools and and so it turns out there's one group of people other than architects there's one group of people that consistently build taller towers than all the rest of us and does anyone want to guess what that group is if you know don't tell me there's anyone want to guess it is its kindergarteners so why is that we're sitting there we're adults we've all been trained in various you know technical things but how come we can't beat kindergartners in this so it turns out kindergarteners don't do that whole posturing around the table for 15 minutes trying to see who's going to get

the idea they go straight to building and their Tower Falls and they build again and it falls we only allowed ourselves one tah-tah moment at the very end of the 18 minutes and kindergarteners on average allowed five and I'm sure we've all people who've worked on games have been on that game where the games kind of not there and there's about a month till ship and you have that last at all moment and hopefully you can get it playable so this is the essence of prototyping and so to give the formal definition of prototyping you know it's a early sample it's built to test a concept or a process and it's actually something to be learned from prototyping is not for the consumer down the road it's for the team Steve Jobs and crew would never think to sell this prototype to the public to make it polished and clean something for later but you need that prototype process and I think somehow we as adults and we as game developers got off track you know the kids naturally

iterated no one ever had to tell them there's value to it and in my experience many of the game developers we didn't and so how did we get off track so I look back I've been working on games like I said for 20 years one of the first games I worked on was a game called genome if you're one of the ten people that bought it raise your hand but back then game developers were more generalists and the specs were really low like maybe 250 polygon characters maybe a 256 by 256 texture map so we got assets done really quickly so it kind of got the rapid part of rapid prototyping down but we were we were not mature developers back then the minute we got the first thing in the game and it worked we're good we moved on to the next so fast-forward to Gears of War 3 we're now matured developers but triple-a you know next-gen content takes so long to create and so from that

moment someone wanted to design something and then see it implemented in game for instance something like a boss battle it was now twelve weeks versus what would have been two weeks long ago and on Gears of War we were mature enough if we got something in say a boss battle and it wasn't working we were mature enough to tear down some stuff and rebuild it but at most we got to tell all moments it was a very expensive process Gears of War was not known for its great boss battles and that could have benefited from a more formal prototype process we just didn't have it so as game developers you know what is our marshmallow challenge like we're not just building the simple kind of objective goal thing like the tower games are really complex things and they have a lot of complex systems that work together you know you're gonna get a much larger team than four guys around a table and they're made for an audience the audience is ultimately is what

judges your success and then games require a creative process so how do you quantify a creative process so I just want to let you guys know to talk about prototyping I kind of want to give the whole context of what prototyping happens in so I'm going to like zoom really far out and talk about some things that we probably just all take for granted and that is what is the creative process like can we parse that down to something describable what is a game we all work on games but what makes a game different than something else like cooking dinner or riding a bike and then what is our role as animators in this whole thing so the creative process I don't know if you guys have read this book by Scott McCloud we had this one chapter on the creative process that was really fitting for this and it goes something like this you have the idea purpose and that kind of comes from just that initial inspiration it comes actually comes from

part of the brain that doesn't even process language is that initial inspiration impulse but you do want to get that out to the world in a form of something so as a creative you're gonna figure out the form then the idiom that's kind of like a more narrowly defined version of that form that's kinda like the genre you know so if you're making a movie it's gonna be a horror movie it's kind of like a genre and then past that you have a structure and that's all the components and how they work together to achieve the form that you're creating and then you have the craft and that's why a lot of us are here today as animators is to learn how to better our craft and this is where we put a lot of the blood sweat and tears into what we do and then you have the surface and this is what the audience sees you know they don't see all the stuff that happened prior to the surface they just see that on the shelf and this is like a good way to kind of parse the creative process and oddly enough all of us in here we kind of we master this

creative process backwards we all started out on the other side of surface we can we started out as a consumer seeing the surface of someone else's creative process but something in us made us want to peek behind a curtain maybe it's like a making-of or an article you read but somehow you got to the side of craft and that's what brought us here today so I'll put this in context and make any game so I'm gonna use Cliff Bleszinski the maker of a Gears of War and run through this like how it maybe it went like this I don't know for sure well let's say cliff and his brother they're sitting there they're talking about when they were kids and they would go out to the you know out to play and explore and cliff want to make a game that kind of encompass at that initial motivation the the inspiration to make something and they're remembering that and he's going

to put it into a form and of course that's going to be the video game he's not gonna make a board game he's not gonna make a movie he's gonna make a video game and then to create this the idiom it's gonna be a 360 game its third person he wants multiplayer and co-op so he and his brother can play and then you get to the actual game itself and a game is its structure that's all the components within it and so for Gears some of the systems were like running shooting mechanics cover all just the little systems you need for your game text to be the game and then of course it's a craft this is the very slow part of any creative process because this takes like the heavy lifting work right here and then of course the surface so Gears of War is released it's a final product it's on the store shelves everyone's seeing it so did cliffs initial inspiration survive all the way through to what the audience got it's kind of like the skin

of an apple you know the shininess is there but the consumer is going to take it and they're going to bite into that Apple and are they going to be pleased with it and a lot of games have that beautiful shiny marketing touch they have beautiful craftspeople the craft in the game is really good great renderer great models but then you play the game for a couple hours and you're just not into it and there's a bargain bin full of these and so how do we how do we avoid that like how do we get you know we put so much effort into the craft how can we make sure the games we're working on aren't gonna get to that bargain bin so let's go back to structure this is where this is where you could ultimately fix it cuz I think this is where it breaks breaks down and so we do take this for granted working on games but what is a game what is what it's like a core structure of a game that sets it apart from any other kind of activity we do and it goes something like this I got

this from stone the brandy a guy who works on a designer at EA so all games start with a starting point and they have a goal and then there's opposition to that goal but you allow the player decisions and tools to get around that goal and all of this is encompassed within the boundary of rules and then how the player interacts with the whole thing and each of these components is something to be dialed with depending on what kind of game you want you're going to tweak the decisions the opposition and all that so just show some other examples we all know this one starting point goal opposition tools and decisions and the rules of the game and it's not just video games you got your kickoff your goal your

opposition your decisions and your rules so once you kind of like have the game down down to that you kind of know where what you're looking for when discussing the vernacular of a game so where do animators come into this scenario so I'm zooming back in now to animation and like film animators film animators serve the purpose of telling the director story to help communicate the story that's what you guys are for video games you're communicating the game you're communicating I hit a button did the character reactor communicating if that was successful you're communicating if you broke the rule somehow or if you were successful with the decision and it's what we do and we share this task with audio effects you I there's many ways we communicate to the player about the state of the game and the game designers kind of orchestrate this like an orchestra like sometimes animators

have their solo as the key communicator of the game sometimes we all share in it and this point was brought players to me so we're starting prototyping on a game and the designers told me like don't worry we don't need animation we don't need anything we're just going to use rectangular cubes you know in place the characters and so they have these cubes running changing speeds and direction and hitting backwards if you hit him but it didn't communicate they finally came and said look we need we need something more than this this is just too too raw to like we cannot tell and we put in bipeds and suddenly the communication of the state of the game was like instantaneous and this kind of really solidified that role for us animators what we do and of course this is where we so this when I'm talking about prototyping for games I'm not talking about craft you know sometimes in crafts we do a bunch of stub animations to kind of get the

designers off our back so we could spend more time polishing the animation well that is a part of prototyping this is kind of the think of the structure side versus the craft side of prototyping so if you're on a project like mine where it's day one it's not a sequel you don't have a ton of resources to draw from what does day one look like and this is you get your designers for their gameplay pillars you probably have some tech in place already and they're doing gray box levels and the team's just ramping up the art team has nothing there's no models yet at best they're scribbling some concept but we don't much so as animators how can we help designers instantly and so that was our challenge so we wanted to make a pipeline for prototyping that was still useful once we got to production but it allowed us with them it allowed some tools and so our goals were to make fast

prototype assets like make it fast but not ugly I mean if people are gonna be looking at a prototype game for a while you don't want any distractions and ugly temp art really just grinds on you today after day and in a prototype phase since you are working on the game with those components communication is all you're dealing with you're not dealing with showing how great of an animator you are at this point you want to just communicate the actual ideas being done and you're not going to noodle on just fine polished animation and then there's no rules is there's no shame in reusing and retargeting run cycles over and over and over so these are our goals here a little bit of our goals and so we made this content creation tool for Maya it's called the auto rigger and what it does is it creates a fast prototype of the character let me get this playing here it's a plane

alright so what this does is it allows us to create a biped really quickly and we can scale different various portions and it's done quite quickly and then we can just snap it to an auto rig that gives us all the controls with AI kfk switching spline I K we could load animations onto it load mocap export whatever we need to do but the key is to do it fast and so this character here I think in the end took about 45 minutes to before we can have it import it into the editor and in addition to that we also have kind of this real simple texture for it where we kind of go into Photoshop and just Paint Bucket in the various points because in the game sometimes you want to communicate who the enemies are who the team members are maybe what buffs would look like or anything else and so we have just a quick color scheme and then so we have

this guy and what we did was sit in a room with a designer and we're asking him like what what's your idea for this game and as he's talking he's describing the verbs and so I was an animator you're gonna sit there and you're pulling out the verbs from his game design way he's talking about the idols so I'll say the guy's gonna punch and he jumps and when you shoot him he falls backwards all these things are verbs to be communicated but it's not just communicating verbs that you got to think about who you're communicating to so sometimes some of the animations on your player is for the player himself the guy with the controller but sometimes the verbs are for the opponent so you gotta keep in mind who these words were meant for um with Gears of War we noticed that a lot of times we were spending a lot of time polishing animations but no one was watching Marcus they were watching the reticle and screen and so then we put more effort into the animations that we knew people were looking at and as often time the

times the opponents and it is said that with video games you have to make the animations look good from all angles and that's very true when we get down to it there are a couple of primary angles you're gonna see of the character in the game and just knowing who you're communicating to in the game whether it's the opponent or player that gives you a sense of does that pose need to look good from the front the back you know what's the camera angle in the game and then we have we're developing Unreal Engine 4 right now and we have a really strong focus on animation tools that are animated friendly we kind of got tired of in the past creating just this big list of animations to kind of throw over a fence to programmers to get in and so we have these these tools called blueprints that basically we could take the variables that the game has like say speed height you know running and animators can go in and plug anything

they want into these variables so let's say the ponds what we call it the pond is running forward at 200 units we know we could trigger the run cycle and so we have things like the state machines but the emphasis here is that it allows the animators just to drag and drop the stuff they need and we work with the rule set at first and of course that's what the input of the designer but that allows us to mock everything thing up so within two days easily we've created a character that's pretty much decoupled from the art department we've created a character rigged it put in rough poses animations brought it into the editor and we've given it its basic run cycle of basic behavior and we passed that off to the designer so you can start prototyping and iterating and end up being like super lightning fast so sadly

I can't show you guys the game I'm working on right now we're in the prototype phase but just kind of give another example this is that same prototype character I did earlier and another one of our tools is just the simple aim offset that's additive layering and we made it very simple for animators just go in and drag and drop this stuff in and so we have things like blend spaces so depending on the angle and running we can then trigger other animations and all these tools have really put a lot of power I guess into the the animators hands but it's also made us like it's definitely had it made us put a foot into the design aspect of developing a game and so as we do this we you know like I said within let's just call it a week within a few days we now have this thing in we're ready to play tests you know what the designers want to prototype and we

found that when we first were testing you know these these mini games we're prototyping we found that if something wasn't fun we would go back to our desks we had a list of you know sometimes a dozen things and we all just go back and tweak things and then the next day the game might be fine but we didn't know what the magic thing was that we changed to do it so we kind of had to limit ourselves to only picking one to two variables between play tests otherwise we're just kind of spinning our wheels in place and then as animators everything we had in there we had to ask ourselves is it communicating you know do do people feel like if they're getting shot do they feel like they're getting hit reactions or if they're getting punched did they see the anticipation so a lot of the times animators are sitting there just making sure like we are communicating and sometimes depending on the game like how far the camera is from something sometimes the animator no matter how strong he makes this pose he's going to

need help from an effects guy you know to kind of put some kind of maybe effects emphasis on something so this is where we we play each day and we kind of just get to that point of where we're communicating the game to the best of our abilities so we know that if the game feels fun or not fun it's it's the game itself and not because we're lacking in communication and of course we test for a responsiveness and of course when play testing nothing is sacred you've got to be in the mindset that everything and anything can be thrown away at any time for something better so so we're prototyping this stuff for weeks on end and then suddenly we start hearing from the art department they actually are making great things well from my earlier slide the art department is very it's very slow to create next-gen art assets so we had to ask ourselves like how do we roll this into this prototype face you know if we now have to wait four or

five weeks maybe six weeks to get a model in do we hold off on prototyping certain aspects and so that was one of the first questions we asked and the other one was how do we just roll some of the artists into that mindset of prototyping and so it we sat down with with the art team and we kind of showed them our prototypes and we said look here's what we're thinking and so magic happened like the first day the concept guy played the the prototype and suddenly he got the essence of what he was going to be designing you know we're testing the pure function of the game not the form at all just just a function but by him knowing the function it definitely informed him of the form and but like any artists are not going to just coldly cold calculated stick to just the form they're gonna had in you know what they're paid to do these concept guys and so what we also found was we would get that concept back and

then suddenly the designers are saying things and the concept work that they never even thought of so it kind of got this cool feedback loop between concept art and game designers where they're kind of bettering each other by going going back and forth so we got the concept art down is feeling really good putting that into our prototype process but mod will still take forever the guys they modelled it out they ZBrush it they make all the normal maps and so we ask them like is there a point guys where we can get something quicker from from you so we could try it out because we've all had that at least in my experience we've gotten models before that totally broke the gameplay mechanic for the model itself did not lend itself to the gameplay it needed to be so if we could find out how we could solve that problem earlier and be fantastic so the modelers said yeah sure we can mock-up the broad-strokes shapes of our models early and hand them to you and

we'd put those on our auto rigger that we use for the earlier mission demands same tool and we would get that model in-game and the questions we ask are you know it does the silhouette of this new model lend itself to the communication it needs needs to do all the if the guy has to run and jump and stretch and grab reach over his head kind of model support that and so we were able to get this like feedback back to the modeler so they can make changes very very quickly and when we rigged this we weren't caring about skin waiting we weren't even caring about textures at this point just a broad stroke stuff but we had that model and the animators were kind of panicked for a little bit we're like do we make a whole new round of stub animations because we've just made some for this Michelin Man but not everything's going to totally apply to the new model and so we said no there's

no way we can just do two rounds of just redoing all these animations so we have a pretty easy retargeting tool in ue4 we just click and within minutes all the animations would come over and so that kept the play test going the mauve we didn't have to wait for a new model update but we knew that model is going to totally break we're not going to look at the silhouettes of those animations and make sure they work so what we do is all those verbs we had in the key poses we would just make a character sheet with this new model and make sure that his silhouette reads from those key things you're communicating those key moves like his Ron's or if he's jumping up in the air is his anticipation poses and then what was great with this is all the departments we are all kind of in sync with what these all these assets are created for in the past with even gears it was just this concept would

kind of go over a fence to modeling that would be passed over a fence to animation and you could honestly never say we were all in sync with each other but now we all know like the why and the purpose of the game and so like I briefly ran through like the prototyping of design and the prototyping of art the rapid prototyping but that's really half of it half of the other half of animation is just a technical aspects of it and you do not want to like halfway through production find out that something you're relying on is broken so if art designed everything to have cloth and you find out halfway through production that cloth will not work you can't have it it's just a huge pain so what we do is we're also prototyping this along with everything else and right here I showed it our character with the trench coat on

but actually what we did we didn't even have a trench coat model we just threw on a cape threw on some apex cloth just to make sure it worked and and we tested other things too like how many characters can you have on screen your bone counts and we did all this kind of still decoupled from production art and it doesn't predict everything but it lets you get a better idea early on but there is a harsh reality - this is so epic we're self-funded so we don't have to worry about the nervous publisher saying show me something pretty or we're not going to greenlight your game and that's a that's a true reality I know we've all done stuff animations and to us it's clear their temp animations but someone inevitably comes up and says hey man is that are you going to redo that animation at some point and so there's always this communication of prototyping and so you don't want to give up

prototyping - you know please those that don't have the imagination to see what game you're making so don't skip prototyping but since you do have the art direct the art team make it make that a part of prototyping make them make all the visual assets to kind of fill in the blanks for people I don't have said I'm in a situation where we're not dealing with publishers I don't have the end-all be-all answer to that but I think if you make a visual prototype on the art side that's going to help that way you could have kind of your gray box prototype level going and so the developers can still learn and fail quickly what they need - and I've said this earlier is you want to make the Tim part appealing and you just don't want to grab a bunch of just assets from here and there where nothing feels like it's part of the same thing and because you're gonna be look in prototyping you don't want any distractions you just want to get to the core of the game

and so when it comes down to communication of prototyping you know the film industry they got this figured out you can show a storyboard to someone and they know what you're getting at they know that that you're prototyping the structure of your your piece through the storyboard and so we had to make this little demo for GDC last year and we had a very short amount of time to do it and we had to kind of limit our risk so yeah we've just kind of scratched on post-it notes to kind of find what our problem areas are so this is totally acceptable and people didn't panic on this but I feel like with games we haven't quite got to that point where people are fine with seeing a prototype I think so I think part of it is is the more we do this as an industry the more will be an acceptable form of

prototyping like case in point no one's gonna walk up to you and say hey man is this final guy gonna have a scribble on his face you know but we do get those kinds of questions with game prototyping and there are some kind of realities to the team dynamics of a prototyping team is we've all spent so much time honing our craft that it's hard to jump on a project where you're asked to make stuff very quickly and throw it away and it takes a certain type of team where you do not want to keep you don't want the team's large I kind of want to keep it small so it kind of keeps the communication loop tight but you also don't have too many cooks in the kitchen and since its rapid prototyping it actually doesn't take a large team to create a ton of assets for it like I said earlier I used generalist who are excited by tinkering and who do a little

bit more beyond their you know their specialty and with us there's definitely a personality type for it we had some people on the prototype team that just begged to get off of it later because they they just want to get back to the craft part there's other people that just love that tinkering and trial and error so I think when building a prototype team it's important to find the right people for it and then what's to be gained so we were able to like fail fast fell often but there's other things to be gained for the project going forward and that is you have all your systems at least a good idea of your systems in place you know that you're not going to get some kind of feature creep later on in the game and so you kind of have some of your production cost and needs you kind of know what your animation by just going to be in the future without ever having to jump into it headfirst and then of

course during prototype we don't have everything we need and the prototype period is relatively short so we don't have time to build new engine features so the things we do need we can inform the engine team of what they need to add going forward when we go into production and if you don't have an engine team there's some there's middleware other software to help you out and then also back to that communication thing we found that kind of based on how you communicating the game kind of is going to define your style so with animation if the games depending more on animation to communicate you're going to need a style that is going to allow you to show more stronger poses like a subtle mocap style might not be able to communicate everything you just need to kind of choose your style guide maybe inform the style guide a little bit more based on what needs to be communicated and then

do you have enough information to jump into production and you're never going to know everything before going into production there's always a plethora of unknowns but if you can just cut down a lot of it during prototype its it helps out a lot and then so with gears one we had say a ton of stretch goals and a prototype team's gonna have a lot of stretch goals and if you get them in great if you don't then if you do work on a sequel it's gonna be easy you're gonna have so many like you're already going to have these systems who are totally motivated to go after and so Gears of War 2 that that was nothing but Gears of War stretch goals that we weren't able to get to as five gears of with Gears of War 1 and then as an animation director being on the mocap floor and directing the actors I kind of knew what the

communications for so rather than just being on the floor and watching their their movement from all angles I know the gist of what they have to communicate strongly and just knowing the structure of that does what they have to communicate just allows me to give much more much more focused feedback to the actor and then even to the guys on my team when I first became an animation lead I tended to direct them on the craft the the 12 principles of animation in the craft and enough stuff but when you're solidly aware of the structure of the game and what you need to communicate you're going to give more a more applicable feedback to them and so they can make decisions that are that are not just aesthetic but more functional for the game and then so asset creation was less like the

telephone game and so it's more like we're in a room together between all disciplines so the modeler leads modeling leads the texture leads effects leads we all just kind of were on the same page versus using an old old system of throwing assets over a fence and so like I said earlier with this with the the creation or with the creative process we all started off and for some reason many of us here decided to peek between the carton between surface and craft and we stayed in craft for a long time and I was thinking about films and I know several animators who went on from the craft of animating to become a cinematic director and we got we got film directors like Disney Tim Burton Lasseter and Brad Bird so they all started off as animators they ended up as directors but I thought about with games I don't know of very

many examples of the animator kind of going to the path of a designer and what if the reason is I think old tool sets and old team structures then allow for the animators really to peek into the structure of the game as much we kind of were just making animation lists and passing it on for someone else to plug in and so with this prototype process even on our team I noticed that animators were suddenly being more aware they're more aware of the what they need to communicate in the game and case in point one of our animators just graduated on to the design team because of this he found that he he was he liked the structure part the designing part but you got to transition there via animation which is a first for me that first for me to see in the industry and or any companies I've worked at and someone like me I like craft I want to stay I want to be that animator but I know having a one-foot knowing the

structure of the game I'm working on it just allows me to ultimately be in a better craft person at animation and so we're back to our roots you know back several years ago it took two weeks with this prototype process we're back down to one week and and and but we had the maturity we had you know we know when we need to go back and keep improving something and then so let's get back to surface it's the final total your games out there's no guarantee that your game is going to completely succeed but you have allowed your team many more times to fail and try again and with a good prototype system in place you've allowed that and you've allowed yourself to kind of get back to that thing we kind of always knew as kids and it kind of ultimately

allows you to be far more creative just get to do it in a faster pace so that kind of summarizes the prototype prototyping for games so if you guys have questions feel free and thank you [Applause] and I want to thank these guys for the reference yes mm-hm

yeah so the question was are there tools out there for a smaller team to use if they don't have their own tools programmers or a TD to set up the system I believe there are I mean I know mix mo offers Anam sets for people to jump off with I know our tool we use for Maya we based a lot of those features off the is it biped for max you know a lot of that so a lot of its there if you don't mind it being very gray box looking I'm sure there are a lot of options out there yes yeah so in this prototype process actually being in Maya was the fastest

part because really we're just getting the assets out as quickly as possible so we found ourselves in the editor more I'm just kind of setting up the rule set and the flow of kind of like the logic between all the animations and how they blend into one another yes mm-hmm right that yes I that's one thing I probably left off is he well first of all you

have to have just good communicators on your team one thing we did to make sure that happens we would sit in close proximity with the designers and we kind of had a animator and designer kind of pair up and after like the first couple iterations they kind of learned there how to communicate with each other and work together so did that answer your question [Music] it depends the personality type there's some guys who it and we found this I mean it's actually really hard for someone to do a bunch of content and throw it out to be critiqued even when everyone's aware of prototype phase and so some designers kind of like to keep their whole vision like this and others were more than willing to throw theirs out for the play test that day so it just takes time it took I think it took us maybe four to five six weeks to kind

of get that process flowing yeah yes sorry hey so you were talking about how you're communicating the idea to the user as a designer what sort of ways could I'd be able to communicate you know the idea to the animator to kind of increase the productivity on that end that's a good question I mean I'm taking a lot of stuff for granted here because our designers and animators have worked together so long it's a really good question I think I think it's just sharing knowledge between the two I think if designers kind of speak about the the cores that's why I kind of

brought up like what is the game and the components I think the more animators can kind of get that language down the more likely they can kind of have that common vernacular between the two hi so when you were talking about the visual bar in a sense or the quality of the prototype in regards then showing it to other people are you were saying that because you're working internally mostly and not having to show to publishers or other people it's not quite so stressful but I was interested if you could elaborate more on where where that bar ended up being for you you know mean like was there a certain level of a lighting or a certain model quality or a certain animation quality that you ended up finding was the right level that people didn't mind looking at the prototype and could get past the visuals and see the right practice but what was going on behind it yes so the prototyping was just for us right it's just for our team only not to be seen by outside people really and we could

probably communicate with a lot of wording to them but I think it for us it kind of naturally happened like we wanted the aesthetics to kind of be soft on the eyes and kind of have everything United so the shading on the characters need to be the same as the shading and environments so we at the bare essentials it still had to have a little bit of aesthetic this to it but we also wanted to decouple it from art as much as we could but really our goals weren't to communicate outside to people as much is just give us something where we could instantly know if something's working or not and get rid of all the other distractions so you had like animation bingqing and this sort of you know much more simple animations or did you did you do another part of some animation so we generally would do blocking like first pass blocking animations to get it in and get the timing it depends what kind of system you're

working on so let's say if it's something like a melee system we found those needed far more kind of advanced animations to kind of get just a prototype working and then other things like idle poses you don't need all the little subtle you know rub your nose stuff so those words usually like one frame to kind of just embody who the character was and so it kind of is basically what what got it enough to be prototype able you know it's basically the role set for that it was really hard for us to not go beyond it like I found myself just noodling on animations when I didn't need to thank you yeah I'll go to this side so I get to do a ton of prototyping actually at my job and that tool looks awesome and most time when like a prototype requires a bunch of animation it's usually like an immediate turnoff just even because that's usually considered like a huge time sink for a prototype right right so I was curious like the tool looks so awesome did you

actually have kind of the reverse of like designers were it was so intuitive that designers were able to kind of like jump in tweak animations and and like did you guys build off any ideas just strictly from the tool of like making a prototype more animation centric and like manipulating the character yeah I mean actually I don't think we had the problem of say designers kind of coming into our animation tool in Maya and doing anything but when it got to ue4 honestly right now we're trying to figure out where the line is between animator and designer because the tools are kind of simple enough to where each one can kind of just go either way the animator can take design really far or the designer can go and tweak animation stuff and right now we're kind of keeping that open because we're trying to figure out which which way is gonna break and which way it works most effectively so I don't have an answer for we're actually kind of that's another thing we're prototyping is the where does it end and where does one begin we definitely found

a lot of kind of and but that's why it's important for the designer and animator to just kind of team up and work together so that way one doesn't know that what the other one tweaks are you seeing more of a trend where like some of those tasks are becoming like kind of a one-man thing you know what I mean like you kind of have like a designer that's focused strictly on the animation portion actually in the past we needed that we needed designers too we had to find a designer that had a good eye for animation but I think with with our new tools our focus is to allow the animators go in and there's some core things you can't change like sometimes like some of the timing of actual events and but a lot of the stuff like blends and transitions and stuff are kind of becoming more open to the animator and and like that one animator I talked about who is now over in the design team he kind of was the animator that kept pushing into design and he actually was

coming up with stuff that the designers liked and they begged for him to be on their team so like this this process kind of blurs some of the lines we say it a lot that we're kind of stopping referring to ourselves is like the animator the modeler we're all it's kind of back to the game developer the game developer who animates a game developer who can you know model and really suspend the the tools they're you know we're kind of cutting down some of the barriers between departments cool thanks yeah hey I just wanted to make a quick comment you mentioned animators who became designers and such they didn't know of any actor you know one who was Fumito who ADA the lead of Eko and Shadow of the Colossus so there's the one guy oh yeah yeah and and they make apparently good games anything wrong son canceled so I'm aware

do you think a team of prototyping group for a company can that benefit from like generalists that can do multiple things that I'm sorry I can't hear I can't hear that do you think a team and in a company of that's especially designed for Bertha typing can benefit from like members that are generalists I can also do modeling and texturing and animating out same time yeah absolutely we definitely benefited from the guys who are more willing to be generalist and go out and if they don't know something the guys that go out and kind of teach it to themselves rather than just wait for that I got the specialty to come in so it's definitely geared towards generalist and in that way you're not you can keep your like beautiful that the animator who can just create the best quality animation you're not making him do this he could still keep him on the production team so do you think in the future maybe one

company has implement like for the Taiping themes for games do you think this will create like opportunities for journalists or do you think it will always be specialized and it depends for the game industry is going there's so many types of games that I think there always be a game company who would benefit from a generalist more than a specialist but there's other games that need the specialist I don't know of any game industries in such a change right now that I don't know where the trend is heading I know for personally at epic we we used to really want to specialist animators now we're certainly looking at the generalist who and when I say generous I just mean people who don't say hey that's beyond that's beyond my role you know it's out of Maya I don't I don't want to look at it we want the people who are willing to dig into a game engine and figure stuff out okay thank you

so that was yeah that was a great talk I'm I work at criterion and it's really interesting so we've been going through a really parallel what you guys have been doing and come to a lot of the same sorts of conclusions but I was just wondering was this one prototype contained that was really focused on the animation side of things or was it across the whole game experience that was being prototyped this prototype team was for the whole game okay and with two animators thrown into it and we learned a lot so so my talk was kind of on both straddling both ends of it but yeah it's a and it's new for us we took a bunch of guys who are used to these are guys that worked on Gears of War were used to the production Paulette the craft section of making a game and we we always blasted through prototyping stuff I think in the past prototyping was always done maybe on a per-person level or maybe a couple people would prototype something over here but not a effort by the whole team

yeah that's that's sort of being exactly the same for us over the last few projects yeah so but how far are you guys taking the prototype I'm just really curious to know um to what extent you sort of fleshing out a whole game experience as like a um yeah it's a good question so the the fell fast fell often philosophy is good but only if you have really well-rounded goals yeah I'm so I think whatever your goals are that you need to prototype I think it just needs to be worked out some we have a little bit of extra time to prototype not every company is gonna have as much time you might only have a couple weeks so you're just gonna write down you know you just got to figure out what are their risky things that you want to try out and prototype so for us it's certainly I would say it's more of the structure of the full gaming experience we're trying to find like what what what like gameplay loops what kind of you know structural elements will help make this game so I got time for one more question

my question is regarding AI K is it something that you considering when prototyping and is it if it's not through during prototyping is it a programmer task or an animator concern or responsibility and finally do you thing for the AI K system in Unreal Engine 4 to replace human eye care at one point so yeah we don't have human ikd integrated or at least at epoch we don't but I didn't even mention on my slide I was showing okay that in the prototype phase where the animators are setting that up because we're all totally the ones that kind of need to define when and where it's used and what it's going to be best used for so we're kind of handling I guess the first implementation of I K but if we get to a section where we need like predictive AI K we're just going to pull in a coder to help us with that system alright great questions guys thank you so much [Applause]