Animation Isn't the Answer

[Music] it's kind of ridiculous over-the-top clickbait t title but that's what we're kind of going with here it's one of these things we're now over multiple innovation boot camps and at the Animation roundtables a lot of the conversations have been about animators themselves becoming a lot more technical having more ownership of becoming technical and we wanted to sort of touch on the next part of that step where if you do decide that as an animator you want to embrace becoming more having more ownership on implementation of your animations in the game what sort of conversations is that going to lead to what are some general personality types that you're going to run into as far as the other developers and disciplines on the team that you need to sort of be aware of right and how can we sort of maybe help get you past some of those initial speed bumps

so that you can get right into the fun of maybe the next level of implementation and and get through some of those growing pains but that's pretty straight forward so we're gonna play it up a little bit what we're gonna do is we're gonna have this a street esteemed panel of amazing engineers and animators play some very specific roles that you might run into when you have these conversations so first up we're going to have the tech enthusiastic animator clay Mathews [Applause] listen I've worked on Red Faction gorilla the Saints Row series and agents of mayhem but my claim to fame I guess

is that I don't elbow drop animators when they complain about wanting to be not wanting to be technical so yeah excellent alright and next up we're going to have jeff horror our cynical animation programmer thank you everyone Jeff here we're here I am principal software engineer at Electronic Arts I've been there since 2001 worked on pretty much every EA Sports title in some fashion my claim to fame is probably back in the day I was one of the first engineers working on our in an internal animation tool ant and my other claim to fame is I've choked out two animators on this very stage at GDC boot camp in the past I'm currently a gameplay lead on UFC 3 and I've got a animation talk on Thursday morning at 10 o'clock check it out yes alright next up on our panel is going to

be the tech futurist Paul mikanos I've worked on many many sports titles just like Jeff I've worked on a photo some non sports idol - like battlefront and so on my claim to fame might be that I wrote a bug that was so popular in the British tabloids it was featured and had over 2 million hits on YouTube and it was fantastic you can look at two soccer players kissing each other it's it's fabulous and then we have our devil's advocate - all of these specific viewpoints mr. Cote level hello my name is on the gameplay programmer at Blizzard Entertainment and my claim to fame is that I've helped build and design animation systems for heroes of the storm and generally stop animators from getting too technical but people talk about that all right and then I will be playing in

this the role of the unsure animator who is breaking out into tech for the first time I'm Mike John Bluth I'm a senior animator at Bioware all right so all right so let's get it started the design of the game has been said that we need to have better melee combat so as an animator I guess that should mean that all I'm gonna do is just care and spend all my time about making some really sweet synced interactions between two characters and like that's that's it right like all I need to do is just do some of those sweet takedown animations like we saw in the mordoor talk right clay tell me I'm right you're you're part of the way there right the the end goal for the system is a you know is a mate you know it's it's a melee system right you're talking about making the the pieces that that go into that system so if we if we think about it as a you

know as a puzzle you know each of the pieces has to you know has to fit together in the right way very precise way right so if you if you become more technical you can build that you know complete that puzzle on your own and you know once once you have those skills you know if those if your toolset allows for it you know you can you can be completing all kinds of puzzles man and you know we don't we won't need the rest of these guys it can just be me and you so we can make our own game yeah with you know hookers and blackjack and you know like blackjack so yeah just you know let's let's you know getting those skills to complete your own puzzles is is so powerful and you want that so that I don't have to do with programmers done okay cool okay I I don't agree with anything you said I think I think it's cute that you're trying to learn programming and everything but what I'd really like you to do is go off and

build some some previous work for us show us how this could really look good in like the best-case scenario and then we'll just throw that work away and we'll come up with a design that kind of constrains your animations tell you exactly what you need to do and then you'll just start pumping out assets for us exactly as as we asked you to you can learn tech stuff on the side I think that'd be great for your own personal development but we're building a triple-a title here if we're gonna sell millions of copies based solely on this new brand new melee system we can't have a hobbyist building the the architecture for it sorry okay so just keep doing what I've been doing for years but I still get to make a super sweet like previz animation so clays wrong you're right so I kind of like what clay saying and I kind of like what Jeff's saying but I don't like the either of these approaches so what I'm gonna say to you is we're gonna I've

read about this amazing stuff okay and we're gonna make this amazing black box thing you're not gonna get it for about two weeks before okay and what you're gonna do is you gonna feed in all of these animations and it's just gonna magically do everything you need for it and you're like how do I control this well I'm gonna give you two dials I'm gonna give you dial one it's gonna do something I'm not gonna I'm not sure exactly what that dial does and dial two is gonna do something else and they might fight each other or they might not who knows and with that you're gonna be able to take all this amazing content that you've created and it's gonna output the most amazing melee system that's an a pretty good it's gonna be pretty good and you're gonna be able to ship a game with it sweet all right so then I just need to make some animations the way I want to make them put them in your thing don't need to do extra in by you don't have to keep doing the same old thing that you do I like it right this is what we're doing no I don't even know it Clay's talking about a black box

says enough I think right there there's two knobs I think the right approach here maybe is to listen to none of these people and kind of choose the approach that fits what you're trying to do maybe a little bit of a sprinklin of both maybe some new tech maybe a little bit of handcrafted animation maybe some new systems so make the handcrafted animation shine maybe a collaborative approach that could I mean this is the hot take I haven't thought about it okay all right so let me let me let me see if I can sort of figure this stuff out then so all right if I am going to care more about the melee system as a whole in some way kind of like what you're saying you didn't call me an imbecile right then I guess really what I need to think about is all right let's I guess all the alignment orientation and timing those are like animation things so maybe all of that stuff should then be owned and implemented solely by me that you still get to do some stuff

but maybe that's the right stuff right like that should just all be that's us that's the role of animator we don't need programmers for that as an animator as an artist you you know you want to own everything it can be exactly the vision you have for it if you learn the tools and how to set it up on your own right like that's the dream I want to build the system that I want to build that's and if I have the skill set to do that why do I need to talk to these other people this is my system I'm the only one required to have input on this right but like how many years into development do you start animating am i right cuz like Jeff yeah so okay here's the deal we have a team of what 10 engineers that I went to school to work on these complex problems the architecture and stuff and we're currently bottlenecked on animation I don't want you wasting

your time on anything that's not involving animation okay you've got a whole team of people that are experts on this stuff we're gonna build the architecture and we'll tell you exactly what we need from you we are a little tight in terms of like manpower so we are also going to need you to do a lot of data entry on things that kind of ties into animation a little bit but it isn't really animating and we're gonna give you some really really bad pipelines and you guys in order to do that stuff and we're not really going to explain to you how it works we're just going to tell you exactly what you need to do and you're going to do it have to do it over and over and over again like a thousand times and design is going to change their mind like probably two or three times throughout production will rewrite these pipelines but we do not have the time to build you eyes or animator friendly workflows for any of this stuff but we really really appreciate the fact that you will take on these data entry tasks learn these pipelines by yourself and not bother us because we're busy building the important stuff like doing like aren't

you sick and tired of these guys who've made these games in these same ways and they're sort of stuck in their box and they don't want to think anything outside of it and they're like you know we have to do it this way because that's the way I ship the game with before we can't go anything further because you know this isn't gonna work and that's not gonna work surely you want to take a leap of faith here and do something a little more pushing the boundary yeah yeah yeah yeah okay so why don't you come and we'll work together and we'll make something that's gonna work most of the time ninety percent of time the hundred percent of the time you academic types are all the same you release some sort of tech demo that gets the public all excited and my manager is all excited and like hey let's do this machine learning thing for the next project and it never works in practice you just give us these cool black boxes that we have no idea how to use we've got a we're gonna stick with the traditional stuff that works until someone's put in the effort to prove the stuff out in the real world and I don't want to be that person I don't want to be your lab rat all right Cody I'm like really confused at the role of animator versus

programmer can you tell me how all these guys are wrong yeah we should work together and flex our skill sets like you could push the boundaries of animation and awesome and interesting ways without going wholesale and jumping into like a pit of knives which is what a black box that makes 90 percent the animations procedurally is right like what a nightmare but you don't also have like sit in a corner and be the programmer and the animator and the artist and the producer like let's me I'm better at programming there's no offense it's all I ever do let's work together let's build it together I'll build a little bit of a box you can have a little bit more of the box and we'll make a awesome animation going forward you don't have to do any of these single like crazy OneNote like through and you could I'm saying I do so alright so I guess then if I'm gonna do a little bit more programming work then you guys are gonna learn a little animation work I mean I kind of have to to help build you the cool things you need to animate right okay you know that makes sense alright so if I'm okay alright okay so then I guess if if we're

gonna start building stuff we're gonna start taking a little bit more ownership of this stuff but we're not going to own everything cuz we're gonna work together right my then I guess we just focus on making that current design that we get from the designers and whatever we come up with right off the bat we just do that and it's done and then we can just get back to animating right that's it that's all we got to do we just make that quick little thing and then we're good no you know there's you know any any process is iterative right but you know having a owning the vision is and understanding the vision is critical to to you know ensuring that it gets come you know completed in the right way a singular person can if a singular person can own and implement that vision then you know

it's going to you know you know it's going to be the one that was originally proposed right who ever wanted the the melee system had a vision for it if if they trust you the guy that's setting it up and you can own it and you can be the you know the end-all be-all for that system then why not go for that I'll tell you why not to go for that I've seen this happen too many times a tech happy animator starts building some some fancy layering stuff on top of his base animations and makes the animation look more organic and the transitions look smoother and everyone's thrilled about it and they starts to get his tendrils into other things that start to affect gameplay and introduce a whole bunch of bugs into the systems and come finally who do you think they asked to fix those problems they come to the engineers the animators are off doing I don't know what's one can weed up on the

roof or something well the animate while the engineers are trying to figure out these these bugs and then we're exposed to these huge spaghetti mess of visual programming graphs that I have no idea what's going on I come to you and ask you what's what this is all about and you wrote it three months ago and you have absolutely no idea how it works anymore and it's on me to read through the stuff and trying to figure out what the hell is going on I don't want that to happen ever again designers change the mind all the time I'm not going to trust them either I'm just going to build what I think is the best interpretation of what the designers are asking for that covers my bases when they do change their mind and you're just gonna do what I ask you to okay these these guys are never gonna work together Paul so you've had some stuff that's out there that doesn't tell me there's some future solution that means these guys can stop belly aching about whatever they're doing and you'd say what we're gonna do is we're gonna go rent out a mocap space and we're just going to organically just capture everything we think we might want and and we're not going to do it in a sort of systemic way we're just going to sort of organically capture all the combat

that we want in the system and we're just gonna rely on the thing to learn what is good okay so I got a cup of it how much is a lot um how long is a piece of string like you know I mean it's gonna be your job to clean up all the data but you know what we do now yeah and eventually we'll have machines to do that for you as well so you cool the rack up to smoke and weed on the roof I am legitimately astounded the nightmare fuel this sounds like I don't even know how to go ahead this like my I've been burned by animators to end by designers and I burned them subsequently as I here's some sweet turning tech and then it didn't turn I've learned that lesson right but I think that the takeaway here is that like yeah you don't go sit in the corner ever I I think my message is gonna be one of the whole time why would you go and do this alone we're working as a team we can build

things as a team and I think I think better off as a team like vision holders and motors don't leave a place in making collaborative designs in collaborative art like what are you this is crazy you guys are like terrified of everything well here's I'll tell you what I'm terrified of right what I'm what I'm completely and utterly terrified of is how the hell am I supposed to balance being like a traditional aesthetics focused animator who cares about like that visual quality like we care about the performance of the character and now I'm having to learn about some form of owning all the things some form of something but I not saying and smoking weed I got to get good at that now you're talking about like lengths of string guys like how do i balance becoming more technical while all the stuff and why even like what the hell how much of it do you want to own

is what it comes down to right like you were you were tasked with building a melee system right you can you can offer the assets to a certain point but when you start plugging them in if it's your job to set them up and you find out that the pieces don't fit then you're gonna have to go back and revise them so that they do fit right and then you have to deal with design iteration with timing changes all that fun stuff you're you're gonna pick up the knowledge whether you want to or not if it's required of you to you know set the system up so why not just take it a little bit further and then just cut out the middleman in terms of iteration and timing and just do it all yourself okay so learn just enough so that I can keep applying the traditional stuff that I know in game you can't lose sight of the engl right the Engel is the the system right like the pieces you build a part of it but in the end it's you know you're

delivering a system okay I think the the traditional aesthetics is the most important part of your job and that's why you were hired was to make sure that the game looks really really good and we're gonna put a lot of constraints on you in that effort and if you're trying to figure out where you want this this technical development to focus on make sure that it's always driving forward those traditional aesthetics in whatever way you feel is necessary DCC tools that help your effort out automation tools that help get around the bad pipelines that we're providing you just make sure your focus is always on those traditional aesthetics don't get distracted by building gameplay mechanics and feeding AI different parameters so it looks smarter and all that kind of stuff just or control mechanics don't worry about any of that stuff just focus on trying to get the aesthetics to to the best place that they can be and in whatever ways learning technology will further that

goal by all means pursue that okay so I have questioned that because I'm trying to take a page from Cody cuz he's the only one that I feel like is not a rage cannon right now where do you need me more as an engineer helping implement or on giving you a lot of assets most of the time it's giving me assets but the one place where your skill sets can really help is evaluating the game and understanding when it doesn't look good and trying to figure out why it doesn't look good and what part of the implementation is causing that because that's a--that's a part that the traditional technical people may not have an eye for and they may be completely blind to it because they've been staring at it for for so long and they're invested in the technical solution so you I want you to look at the game and understand where it doesn't look good and then try and figure out why it doesn't look good that's another really really good use for your technical skills okay so then how am I gonna balance art and tech with the crazy stuff you're talking about

so what I would like from animator is to embrace technology it is not going to take over your job you are going to be instead of creating one creature you're going to be creating ten creatures you're going to have a multiplicative effect on your output but you can't do that if you're worried just about the specific posing of one particular finger on one keyframe of one animation you've got to start thinking more holistically so I'd like you to start really understanding the technology and really understanding what's going wrong when it's not doing the right thing really figuring out how how I can communicate to the engineer how to make this technology obey the 12 principles of the animation how to make it feel like an animation not just a sequence of poses okay it feels like they're starting to come around a little bit Cody yeah I feel like I don't have more to say here other than like I think there's a balance to strike where you get to own some of that implementation because at the end of the day you have the vision for what that animation is gonna look

like I don't like as an engineer building the system for you I am gonna look at it for 14 hours a day and be like that looks like those quaternions are doing their their quaternion and then I'm gonna move on but if you're like wow that Oran looks terrible and I don't see that tell me so then we can work together to make it or conversely if we make a cool system you get to have a hand in making it too but there's a balance to strike I think we're convinced in some of the engineering people to kind of get on the same page okay alright so then what I want to do because I feel like we're starting to get to a point where we're understanding and I'm a little less lost when I'm dealing with some of you guys that like what's your closing argument what is it that you feel like you need to say to me the animator that's starting down this technical path I think you need to keep your eye on the prize you know don't lose sight of the end goal you know you you started out by saying hey I need to I need to make a melee system and you're your first step in the process was all spend all my time making the pieces

right you know that's a you know can't see the force for the trees kind of thing where you're you know you're you're going to low level with it you know make sure that those pieces you know or in service to the to division you know to the to the goal just don't lose sight of the goal so for me it's a little knowledge is a dangerous thing and complexity will sneak up on you I've seen it happen before something that looks like it's a simple problem to solve with some sort of scripting solution slow very very slowly gets more and more complex so just be aware of that fact when you start dipping your toe into this kind of stuff is if it feels like it's getting out of hand you don't think you can control the system yourself anymore or it's it's feeling like this thing isn't gonna scale I'm starting to get bogged

down with with trying to debug this thing because the debugging tools aren't very good whatever it is don't don't feel that you have to take that on yourself and continue to build on that poor framework and increase the complexity go to someone whose job is to manage that complexity and say hey look this is getting too big I think we need to re-implement this in a different way is that something that I can do or is that something that you should be doing and just make sure that you're aware of that stuff don't don't go down this path and get lost in a sea of complexity that you're not quite ready to handle yet cool I guess I would say that if you're going too far and you're like I made this system now I just need to add 12,000 more assets for it in order to be complete maybe you've gone down a path that's a little too structured and if the engineer is saying to you we have to do this in this exact way and we got to do this with this much timing in this thing maybe that's a little too structured so you've got to work with the engineer and work with your own wisdom to say like hey maybe we can create a system that could do this dynamic maybe there's a way that we

could leverage both parts to create something that could be done a little more procedurally maybe I don't need to make slope animations for 12,000 different creatures maybe that's not a good idea and use of my time cool I mean I see the irony now and having this talk we've got soft skills versus like becoming more technical animator but it really is about that right like I agree with what they're saying like we I want to as an engineer I want to enable you to make better and cooler art and that might mean giving you more control or not to be more control or make any more procedural I don't actually know the right answer to that question it's so context dependent but the key takeaway is like collaborate talk to people if you can't do that that's solved that problem and that's a totally separate issue right but like let's work toward making you more technical as a team I think that's the takeaway for me all right so we have a couple minutes left and so what we wanted to do was offer up the chance for anyone out in the audience to ask a question of our panel you have two options one you can ask them to answer earnestly as they essentially just did or if you feel like you've been having

an argument with somebody and you want that extreme counterpoint you can specifically ask your question to one of their personas and they will give you that answer all right so anyone who has questions and wants to do that come on up to the mics also who on the panel if if these are the type of conversations you're interested in having and having more of all of these guys are fantastic people that will help train you up and are excited about doing these types of things more and more who all appear is hiring right now for animators or animation programmers all right so if you're interested in this type of stuff talk to Kody talk to myself talk to clay talk to Paul and Jeff right like this is something where a lot more conversations need to happen and are happening and hopefully this starts you to get a little bit idea of where it can go and what you might encounter so who's got some questions for our panel come on I

want to hear Jeff say some more inflammatory thing hello I guess you can answer this I don't know I I kind of want to see you guys get spicy but I also earn astana know if you see whether or not there is a world where an animator could be strictly like fine art animate or versus like is life now animators must be technical or can you just focus on the art I feel like outsourcers get a chance to focus just on the art if they stay within the metrics that they're provided but you know I'm assuming most of you people want to be more focused developers as opposed to outsourcers focusing on the art is great especially on like cinematics and you know areas where there's no sort of active input

but anytime you you you layer input with you know with control you've you know concessions have to be made right like you know you can make a bunch of really awesome puzzle pieces but if you've got you know I got a puzzle piece for the Avengers and a puzzle piece for Dora the Explorer they don't they don't fit together right like you know changes have to be made and to ensure that those two things go together and sure there's going to be some sacrifices right you know with you know you know there are you know metrics and and you know technical details of asset authoring that you know that are critical to ensuring that it functions well you know together so pure art I think there are opportunities in terms of game play animation you know you know one of the biggest ones we get for for kids coming out of school is you know they have to sacrifice anticipation right it's one of

the first things that go in terms of like game play field but yes and no got another question you want the genuine or spicy genuine okay so you were talking before about having artists meet up with engineers and I'll learn a little bit what you do you learn a little bit what I do I've been in those conversations and I've learned what they do but they don't reciprocate that how how do you teach people how to learn in that sense well accept what you're trying to teach them we might want to plug when is know the who's doing that yeah so at 3:30 we've got at 1:00 about a teaching a team to fish in communication and all that stuff so I want to step on your toes but we do have something coming for that possibly yes but I saw one here you guys opinions well so for me when I got into the

distri the first thing I was has to do was help build animations out for crow me or somebody who turned into a dragon and I they were okay so it doesn't really do this thing that you needed to do it was all Greek I was like cool yeah totally I'm on that man and I went back to my desk it was like I'm not on this I'm very concerned so like for me I think I'm fortunate that I had to go back to the animator be like look I had no idea what you're talking about can we step through really slowly and no dumb Cody but we're gonna get there and I think that was also a factor of production that guy talked to like their lead or I talked to their producer and was like I need time to sit down and learn like that's part of my job so like it sounds really rough to say but like that's a structural like institutional thing make the case for it because like learning that stuff is invaluable for your studio continuing it's awesome okay it's a as everyone talks about the the animator who wants to focus on their art and doesn't want to get technical but the the flip side is completely true you've got engineers that just want to focus on pure engineering and they don't take the time to understand the aesthetics and the importance of like

responsiveness and feel and gameplay balance and all that kind of stuff and the most the most valuable members of your team on your engineering team are going to be the ones that do appreciate the aesthetics of animation and the ones that do understand the nuance of gameplay even if it's just at a surface level and are able to apply that understanding to their work even if they're not the ones responsible for making everything look right if they have an understanding of what does or doesn't look right they can spot problems quicker and respond to them quicker before it gets out and it's a lot more difficult to draw that stuff back in so it applies I believe it applies to every discipline in gaming how you get people there I don't know but I think it's one of the most important things in this industry is having that cross-pollination between all the disciplines and at least having an understanding of it in every single discipline yeah I got a tip for you so if you if you're trying to get an engineer to help you do something better what I would recommend to you is book a

meeting with that engineer get him to sit down with you at your desk and go through your workflow make him really understand what it is that you're day to day and immediately because the engineers gonna get frustrated like this is a waste of time I got I got to do this and he'll be thinking like oh I could do this to improve this I could do this to improve that and it's like you'll come up with tangible items about how you could improve your workflow and engineer will see it because you'll experience that pain face to face all right I know we're talking the other time but Christine has a sly smile on his face and I feel like he wants a spicy one and I want one more spicy to end us off thank you don't don't meet okay I was gonna ask for genuine born yeah we see this one it's for Paul so they say I'm an animator I want to keep embracing my animation roots and I also want control but I also want to embrace the future how are you gonna help me there I need more than two dials can you can you start spicy in the next week yes so I'm with you buddy I want to create

the best animation system that we have available in fact I want to create the best system that's not even available and we're gonna work together and we're gonna see some amazing stuff and and it's gonna look and sound really impressive and then we're gonna choose little snippets of it and we're gonna demo that and then we're gonna we're gonna build a whole feature set around this thing and we're gonna ignore these little problems that you see and you'll be like but I need to do this in this situation and I'm gonna tell you so to be fair like coming out of character a little bit it is very difficult to balance making a complex system easy to understand and controllable it's basically like control plus complexity they they're very difficult to marry and a nice way that it's as simple and easy to get the result you want so I think it's always just about this iterative loop of is constantly trying to get better and working with the animator to see how that workflow is and to see what the shortcomings are and then trying to

really make a system that isn't going to be so problematic for your particular game type so sometimes the system might work really well but it's not going to be for the right for the for a particular style again so we can agree that the animation whatever happens is going to look amazing yes so okay cool so so that basically means that any animation awards and everything come around then the animator gets the credit I think the mocap artist does right all right that was perfect we gotta go thanks to the panel [Applause]