2D Animation for Games: A Primer

[Music] hello i am tyrique plummer i am a game designer and artist i'm going to be talking today about 2d animation which is maybe my second favorite thing in the world if you're interested in animation but don't know where to start or are struggling to get things to look right or if you understand some of the basics but just feel like you could use some tips and applying the principles hopefully this will be a helpful screed i also hope this will be helpful to people who work with animators and other non-artist people who make project scope decisions to provide some common language things to consider things to watch out for when assessing animated objects and assets and some tips for working with uh artists i've worked as an animator in games for about 12 years now on titles like thor god of thunder epic mickey the power of illusion adventure time explore the dungeon because i don't know and cadence of

hyrule crypt of the necrodancer featuring the legend of zelda a lot of games uh with colons followed by long subtitles for some reason i'm also the creator of catacomb kids which is a platformer roguelike that i've been working on for going on 300 years now uh i mostly work in pixel art but i don't think that'll color this talk too much because really animation is animation and the fundamentals don't change all that much from one medium to the next even if specific techniques and tools might and this is going to be a talk largely about the fundamentals of animation as they apply to game art i also preface this by saying these are all just things i've gleaned from my own personal experience working as an animator i'm not like any kind of super authority or anything it's just something that i care deeply about and i have a lot of opinions about what happens to be my second favorite thing that i also happen to do for a living which is to say please also learn from other animators and people who are

better than me there are a lot of them and i'm sure they all have opinions that they think are right too that being the case let's start off with some general considerations uh before you even begin working on art for your game there are some things to consider with regards to 2d art and animation like most aesthetic decisions in life a lot of a lot of it comes down to what you want to communicate what mood you want to impress upon others what story you're trying to tell but you also need to be aware of the kind of game you're trying to make on a technical level and the challenges associated with your medium of choice because any aesthetic decision is going to be a technical one as well if you're going to need to sprite if you're going to need every sprite to have 12 fully animated facing directions and hundreds of kinds of fully rendered equipment that will be layered and animated and mixed and matched i feel like at that point you might be better served by just using 3d assets to save everybody time and migraines

or if your art style is incredibly detailed to the point where drawing a single frame is equivalent to a full day's work maybe be aware of how that's going to impact the quantity of art that can be made and plan around keeping any animation to an absolute bare minimum or opt for a skeletal system where each highly detailed bit needs to be drawn only once or twice but of course that will come with its own set of challenges um or you could simplify the style to better fit within the limitations of traditional frame by frame animation these are all valid options but they're things that need to be decided upon early and trust artists when they tell you what's reasonable and what's not nobody likes working for hours days or weeks to make something that ends up thrown out because the limitations of their medium weren't properly considered early on 2d animation can be very time consuming and if you don't account for this in your design you're going to have a bad time very long crunch filled bad time

so let's say you've made those considerations you've picked a style that works you understand the implications and you have decided you're going to go for it 2d art traditionally animated nice uh so when you're animating when you're breathing life into these still images making them act and react what is it that you're actually doing well animation isn't just make thing look good although that's definitely usually a part of it animation is communication communication of ideas through motion a character's idle animation can communicate volumes about their personality their mood their status in the world the way they interact with objects the way they walk jump the way they flinch when they get hurt these are all means by which to communicate to the viewer information that might take paragraphs to describe otherwise if a picture is worth a thousand words then a sequence of images is like it's kind of like an essay or something i don't know

uh but honestly that communication aspect is true for most animation whether it's for games or not uh communicating personality and individuality through the motion but animating for games has an additional concern which is communicating the game's state not only must you convey the nature of the world and the characters in a broad sense you also have to present a clear message of what's happening right now and why and how the current game state will flow into the next and how it's a logical conclusion of whatever the previous state was and you have to fit all of those matters of character and personality into that context you probably won't have a full solid 60 seconds of animation to convey that are reluctant heroes feeling uncertain of their ability and in over their head you might only have eight frames of a run cycle six frames of an attack animation and a single hurt frame to get across those ideas moreover that attack animation also needs to convey

here's where the hitbox is these are invulnerability frames this is the cooldown time it can be a lot to keep in mind in addition to the meta objective of make thing look good but this communication aspect of animation is one that i see frequently misunderstood or outright ignored by animators who are just starting out and the non-artists who are making requests of us what is the actual relevant information here what is the game state that you're trying to get across what is the character that you're trying to impart good animation is effective communication it doesn't matter how cool that backflip looks if it also takes away the player's ability to accurately read the situation another thing to think about is the fact that motion can be kind of an aesthetic in and of itself like of course the way things move enhances whatever visual style is already present but it is possible to have an art style that looks great in stills but if it's

animated poorly it can really kill the whole vibe it's also possible to have a visual style that looks very simple and unassuming at a glance but that moves with a clarity and precision that elevates it this is specifically not to say that you should go ahead and base your visual style and just absolute trash just because it could in theory look cool in motion although i mean feel free but remember all the stuff i said about keeping in mind the limitations of the medium if you do pursue that uh rather what i mean is that animation is an important pivot of aesthetic uh it's an important pivot point uh within aesthetic that can elevate or diminish other stylistic design uh decisions without being too reliant on them uh honestly this fact is kind of one of my favorite things about animation in general the fact that i can look at for example a child's drawing and envision a reality in which for all the naivete and lack of finesse of the

still image i can envision a reality in which it's animated in a way that lends credibility and intentionality and makes it look good and cool when it moves i can look at this image and imagine the ways that i might animate it in a way that makes sense and has weight and feels right but that still doesn't rely on changing the underlying aesthetic whether it's worth doing or how easy it would be is another matter but my point is just that good art can have the principles of animation applied poorly and what would traditionally be called bad art can have the principles of animation applied well the form that the style takes does not dictate the quality of animation that is possible also i mean it would be pretty cool to have an animated game that looks like a child's drawing someone somebody done that somebody's probably done that uh so with those considerations in mind let's look at some of the more technical

details of what makes good animation first off with a few definitions that i'll use throughout this talk that you should probably understand you may have heard these terms before but in case you haven't covering my bases key frames these are the frames that show the key positions of whatever it is that you're animating the ones that tell the story of what's actually happening just looking at the keyframes you should be able to tell what the action is in betweens are the frames logically that come in between the keyframes the ones that smooth out the key poses into a cohesive fluid motion spacing refers to how much an element moves in between one frame and the next and smears and multiples are techniques to show motion by stretching an object across time or making it appear multiple times within a single frame and then there are the principles of animation there are 12 of them

some guys at disney wrote them down back in the 80s based on work that was done in the decades prior to that they aren't the whole of what makes good animation because art is subjective or whatever but they're an important foundation if you're just getting into animation these principles are the jumping off point for understanding motion and developing an analytical eye as you consume the works of others and if you work with animators these are concepts that may be handy for you to have in order to more effectively communicate your ideas to them or understand what they're talking about sorry but like i said these principles were written by disney dudes and so some of them are a bit less relevant to games animation than they are for film and tv and such and others are just kind of not super relevant to the intent of this talk for instance staging is a matter of composition and scene setup how you guide the viewer's eye that sort of

thing uh which while it can still apply to games might be better explored through level design or something uh and solid drawing is also one of the 12 principles but it basically just means be good at art which isn't the most helpful advice in my mind so we're going to ignore some of these if you really want to get good at animation though i highly recommend actually going and reading about all 12 principles and whatever other resources you can get your grubby little hands on but i'm specifically trying to cover things that might not be touched on in other places and things that are specific to games and a lot has been written about these principles already so we'll check out the seven principles of animation that uh that we care about in this context for the time being so we're going to start with one you might already know about squash and stretch it's when you push a rubber ball down it gets flatter and wider when you pull a rubber band and it gets longer and thinner

it's the quickest easiest way to give something a sense of life and you don't need to limit yourself to only affecting logically squishy things the classic animation example of this is to draw a bouncing ball but adding a bit of squash sorry the classic animation example of this is to draw a bouncing ball but adding a bit of squash on just about anything is an easy way to increase the impact of collisions or some stretch while an object is moving can impart a sense of speed and give it an object obvious directional heading again it's about communication squash and stretch can look good and also tell you what direction an impact has come from or what direction an object is moving in and how fast but it's important very important that you maintain the volume of the object or character mass that gets sucked in along one axis has to spread out on another if you aren't careful about maintaining the subject's volume it quickly stops

reading as squash and stretch and becomes a transforming thing and what was supposed to have impact might end up having less because it changes from a matter of perceiving force uh to a matter of perceiving malleability so be careful with your with with keeping maintaining the mass of the objects that you're animating when you're applying squash and stretch you can see here uh some characters uh character that i did for for epic mickey and another uh that i did for a cadence of hyrule and see impa sort of squashes down before popping up into the air uh and same same for the uh night dude next on the list is [Music] anticipation

this is the pre-action the moment before a thing happens that lets you know what's about to happen that something is about to happen uh usually it's emotion in the opposite direction of whatever is about to follow a slight shift to the right before you move to the left a dip down low before you leap upward uh anticipation can be big big sweeping wind up before an enormous punch a laser that pulls in on itself and shakes like crazy in preparation to fire or like a low twisting crouch where the character tenses like a spring before a super jump uh this is another character from uh epic mickey razoo and there you can see kind of a few different uh moments of anticipation within this so first of all there's the most obvious which is the flourish that leads into the crouching position before charging forward like that crouching position can be considered like in anticipation of moving forward but there are also some other ones that that might be a little

less obvious uh there's a dip before the actual charging like like he's in the crouch pose and then he dips backwards before charging forwards that's anticipation uh lifting like the actual spinning of the sword and putting it in the hand that can be considered like an anticipation you can see it's moving in the opposite sword come in the opposite direction coming up before going down and being held in both hands and there's also a little hop towards the end of the charge uh that sort of prefaces the stop of of of movement where he like hops up before sliding down and crouching a little bit lower but notice how each moment of anticipation is largely in an opposing direction to whatever action it is that follows anticipation can also be very brief if you're making an action game where reflexes matter to any degree you probably don't want your player character to have to wait a half a second before their input comes to fruition no matter how cool it looks unless you're making a soulsborne or

some such uh you might have enough room to add one or two frames of wind up before it starts feeling laggy for many games player actions don't even have any frames of anticipation but they still manage to look good because they still manage to feel good remember animation in games shouldn't override or betray the gameplay it should serve it it should bolster it uh so in a lot of games where you need that snap that instant action sometimes there's no wind up at all on player actions but usually you can fit like a frame or two in there this is impa again from cadence of hyrule doing a bit of wind up before before uh spearing out a little bit of anticipation importantly anticipation not only prefaces what's about to happen but it can also tell players how important what is about to happen is through contrast

frequently contrast is spoken in terms of color and brightness and that sort of thing but there's also contrast in motion if something that was moving is suddenly still or that was still is suddenly moving you'll notice uh large motions draw the eye rapid motions draw the eye and so you want to be absolutely sure that players can see what's about to happen you want to draw their eye their attention to the important thing don't just make the anticipation longer and give them more time to see it because again that's not always going to be an option like it has to be subservient to the gameplay and that's where like all this timing and stuff comes into play but as opposed to making the anticipation last longer you can make it divergent from the norm from whatever from whatever there usually is whatever the thing is that's doing the anticipation that's doing the wind up change its silhouette change its color

make it vibrate make it smaller or bigger tense it like a coiled snake fill it with potential energy give the player's eye something to latch onto and it will so here we have a wiz robe from cadence of hyrule uh you can see there are a number of things going on to sort of give visual attention to this this charge up before the attack um the silhouette changes uh you can see that that for the normal version he's kind of like rounded uh but as he's charging his hat and other elements become much sharper much pointier uh much more aggressive feeling there's also like a bit of vibration that that that kind of motion can can attract the player's eye the silhouette changes completely the the scarf comes off and begins flapping as another point of of interest um so yeah yeah there there are a lot of a lot of different different elements that you can tweak to

uh provide contrast uh between one animation one motion and the next uh next on the list of uh principles of animation we have timing and timing is pretty much what it sounds like it's how long things take to happen it's where the beats are it's the cadence of things if an animation is a sentence that describes an action then timing can be thought of as sort of where the punctuation is this can be very important in game art because game elements especially characters sorry this can be very important in game art because game elements especially characters tend to have fairly discreet states and how animations are paced is frequently tied directly to the transition between and within these states uh timing informs how readily these state changes will be parsed by

the player in active moving changing and visual environment and the timing is also an essential part of conveying character uh to continue the metaphor if timing is the punctuation then it stands to reason that different characters would have different cadences to their speech they have different dialects different patterns uh and so too will they have different cadences in their motion so you can see here two different characters that have the exact same frames the nothing nothing about the actual frames themselves has changed except the timing um you can see on the left the characters tend to have like a more steady pace uh and the ones on the right have like definitive beats to their movement the big demon night guy like steps and pauses after each step it makes it feel a little more weighty while the one on the on the on the left is is a bit more constant and kind of

lacks a bit of uh lacks a bit of focus on any uh particular pose or moment same goes for the hynox again from cadence of hyrule uh where the one on the on the right like very definitively has a few poses that he holds that that hold focus um and there's more weight and more character to to these motions as opposed to just like spinning around and chucking the thing spin around hold build that that that potential energy and then continue onward ease in and ease out also called slow in and slow out are sort of specific applications of timing uh easing in and out is basically just another way of saying that objects have inertia things take a bit of time to get up to full speed or come to a complete stop uh ease

when you ease in you kind of start with your frames spaced closely together and they get further apart as the motion progresses to represent like acceleration you know ease out is the exact opposite something coming coming to a slower uh halt um it's just easing in and easing is essentially a measure of how spacing changes across time it's worth noting that sometimes these elements that that easing is trying to represent uh are handled by actual game logic you know things like inertia acceleration deceleration uh sometimes like that doesn't need to be coded into the the animation itself that you're working with because that's like the thing that the the code is doing is handling all the acceleration and deceleration and inertia for like you know entire entities for characters and that sort of thing uh which brings me to the point that a

little bit of a break from the from the uh 12 principles um these animation principles were written while operating under the assumption that the animator would have complete control over all elements of the scene essentially that there wouldn't be any outside variables actively changing what's happening on screen as it occurs obviously that's not the case with games which isn't to say that these principles aren't still incredibly useful but rather that it's important to understand when and where they apply in the context of of games and alongside game logic and consider where the game logic might be interacting with these principles of acceleration deceleration anticipation all that sort of thing uh not just not just in the way that i mentioned earlier where anticipation on player characters can be short to avoid visual lag and sluggish input like that's that's one example that uh but also

in the sense of like you need to know like what exactly the character is doing in the game when you're animating for it you know like like if if your character if you're animating an attack animation that can be played while the character is either still or moving that's something that you need to know and take into account because if you animate it as if they're standing still they'll be like sliding around whenever you attack and if you animate it as if they're only moving they might look like they have more motion than is reasonable for for attacking standing still uh so and i've i've i've definitely had this happen before where where uh i've animated something that i was pretty pleased with and ended up being a little displeased because it didn't look great in context because the game logic applied motion that i was not expecting to and it was something that i could have

accounted for had i known that that would have been the case anyway let's get back to the principles that was just a brief aside uh the next principle is that of arcs the idea is things tend to move in arcs just as a general observation uh objects like to be in rotation around a focal point and if they aren't then they're usually being acted upon by by some kind of force that that pulls them in a direction or one direction or another uh either by gravity or because of a literal physical attachment limiting directly linear motion like a shoulder how you know your shoulder limits the motion of your elbow which limits the motion of your hand these are all things that move along arcs they're moving at fixed distances between one another uh and uh the pivot points for these arcs can be themselves moving along arcs uh i mean

if you think of it humans we're basically just a collection of pendulums strung together uh so it can help if you're working on an animation and it isn't looking quite right uh to check the arcs and make sure they're moving along and expect a trajectory without any strange jumps or sudden deviations jitters that make the timing or spacing uneven that sort of thing next we have follow-through and overlapping action follow-through is sort of at the opposite end of anticipation if anticipation is communicating what's about to happen then follow-through is communicating what has just happened and how it continues to impact the subject follow-through is the way your arm continues for a bit before settling after throwing a ball or the way a long coat comes to rest a bit later than the rest of you when you stop running nothing ever really stops stone cold when it's finished moving and follow through is a reflection of that

similarly overlapping action is the notion that different elements move at different rates as some parts lead and other parts follow uh this is uh just as an animation i made a few years ago just for fun um not for any particular game but you can see that has a lot of elements of you know follow through in it uh the sleeve the sleeves kind of dragging behind the lantern swinging coming back to the notion of arcs the lanterns swinging in an arc um the the the shorts also kind of kind of lagging behind their legs the flames being kicked up in front of the boots these are all sort of elements of follow-through and overlapping action and i want to say something about how follow-through and overlapping action are important in games specifically because xyz the delayed elements can help clarify information etc and i do think that's true but more often than

not it just communicates that something has happened and it's finished now follow-through can be subtler than other moments of motion as it kind of tends to be a state of energy being lost and things coming to settle or returning to a more neutral state uh but follow-through tends to ground motions and make them look more believable and weighty just classic tool to make thing look good and you can see uh makoto here i did not make this one uh but you can see her her similarly a lot of the cloth like sweeping behind following the the the um motions before kind of leading the eye to an understanding of what has just occurred um overlapping action can also help communicate patterns of motion similarly to how uh stretching a character along the axis of its motion so squash and

stretch can inform as to its heading and speed so too can a lagging scarf or tail or something uh inform to player as to what direction something is moving clarify what's about to happen by showing what has come to pass immediately especially in high-speed games and environments having some kind of visual anchor to demonstrate the recent past can be the difference between parsing a situation and becoming lost in a rapidly changing visual landscape and lastly we have exaggeration it's not literally lastly there's still like five more principles but like i said before we don't care about those so lastly of the ones we're going through is exaggeration uh this one might seem like a bit of an odd one out a lot of the other principles focus on aspects of animation that are grounded in a more physical understanding of the world as viewing physics from the perspective of an artist the gravity

centrifugal force implied by arcs the inertia conveyed by follow through and overlapping action the potential energy generated through anticipation but exaggeration is no less important than any of these exaggeration of both form and motion it can triple the size of a fist mid punch just for that extra impact just for even just a frame you might not see it but you can feel it um break someone's spine beyond logic to get the get them to twist the way you need to for that extra like wind up that extra anticipation you can have the hero do a full front flip if their attack misses because their momentum just carries them into it and they go head over head over heels anticipation is frequently exaggerated just by its very nature because if it weren't it would be easily missed people just don't move that way like if i'm about to do something i'm not gonna have a big motion

you know to show you that i'm about to do something uh and this is kind of the advantage of the medium that we can freely deform and make ridiculous all the things we need to as readily as anything else in order to get our point across if the other principles of animation are the groundwork laid to keep things believable and grounded then exaggeration is the means by which that believety believability is made to matter uh exaggeration can be thought of as almost a modifier that can be added to the other principles to enhance their impact exaggerate to squash and the stretch really push that anticipation fling those arcs like yo-yos i've been pushing this idea of animation as communication which might make it sound kind of utilitarian and technical but that's not what i'm saying if animators are communicators then our communication is poetry we are afforded the ability to play with form and time as much as a poet might play with rhyme and meter now now we're going to talk about some

common mistakes that people make when animating not just in general uh firstly and importantly more frames is not equivalent to better animation uh buttery smooth motion is not a requisite for good animation and in fact adding more frames to smooth things out can sometimes actively harm the idea you're trying to convey by making things seem a little sluggish in addition to creating more work for yourself i sometimes see people who are new to animation trying to smooth over errors of timing and readability by adding more in between frames and i hope i guess that it'll click together at some point once the motion is completely fluid it's not really how it works even worse is the notion that adding more frames is an inherent improvement or that a lack of frames is somehow a sign of poor animation in general uh like somehow the dream is

to have all 2d animation be 60 fully rendered frames a second and then it would be perfect philosopher's stone of animation that's also not really the goal um far more important than having silky smooth frames that detail every moment of an action as having strong readable poses that lead logically from one to another and timing that reinforces the action if you have a frame of someone with a sword raised in preparation and then immediately after a frame of them with their sword having slashed forward that's still animation and if the poses are strong and the read and the action is readable it'll be better than uh uh poorly poorly made uh key frames that that are tied together with a hundred you know in betweens no amount of in-betweens is going to fix poorly timed poorly posed keys say what you will about the animation of the cdi zelda games but their issue is not that they don't have enough frames more often than wanting to add more

frames to animations i make i find myself having overdone it and i need to delete some to add back in some snap to the animation make the motion feel cleaner one of the first professional gigs i ever had was for a game called amour the unicorn keep yet another game with a colon and a subtitle for some reason it's been there since the beginning um and it was a small mobile game i was the only artist uh it was a good early job though because all the animations in the game were like two to four frames uh this is by the way not a measure of my current skill i'm much better now i could do much better under the same constraints um but it was a great early exercise in readability of motion under very strict frame limits and to this day i don't really think any of these any of these animations need to have more frames than they than they do other than other than maybe the bow i think the bow could probably use some

follow through but everything else is two frames four frames it reads you don't need any more than that so if we're not going to add more frames to smooth out the motion how do we ensure that motion reads as smoothly and continuously as we want especially for fast moving uh objects and things oh that brings us to our next common mistake or misconception smears and multiples as mentioned briefly or before smears and multiples are techniques uh used to show the motion of an object across time by distorting it and they are my favorite thing i love them so much i want to use them everywhere they are so fun to draw this is another animation i did for epic mickey and it is my favorite frame that i have ever made somebody pay me to make a game that is nothing but smears please i will do it in a heartbeat

but i think smears are often sadly misunderstood your first inclination upon seeing these smears and multiples might be that they represent speed that they are used as a means to communicate this thing is moving quickly but that's not entirely true smears and multiples despite how strange and abstract they often end up looking are tools to clarify they are the result of an object moving so quickly that a frame needs to contain additional information in order to properly connect the previous with the next smears aren't used to convey speed but rather they are the result of speed uh and this is something that i think a lot of new animes don't really understand because i see a lot of times i see smears and multiples and and and just blur motion blur being used in the exact opposite way an object that is moving slowly enough to be easily read uh is given a streak

behind it that because well that's what fast things have and so if i add what fast thing has to my slow thing i'll make it seem fast that's not really it doesn't follow the logic doesn't work uh because remember it's about communication and clarity of events readability of motion you don't add a smear to indicate speed you add a smear to address the loss of information when something is moving too quickly to follow otherwise and also because they're hilarious to look at and fun to make so you can see here uh the actual context of that smear from the previous slide uh pete is hammering his fist down uh but you can see the middle one that's what it looks like without the without the smear that's the frame in between uh and you can see that the smear is very directly connecting those two very

distant uh points of motion you know trying to fill in that that uh lack of information that void that would otherwise be just lost another thing to be careful of is over animating that is having too much animation in a way that goes beyond communicating the action and instead begins to obfuscate what happens uh or contributes to visual noise and this isn't just a thing that beginners do i've actually mostly seen this from people who are relatively decent animators but maybe get carried away with making a thing move in a fun animated way while losing sight of the context i'm not ragging on anyone specific it's the trap i fall into and on occasion as well an animation can look cool in isolation but when put into context you might suddenly realize wow that animation that idle animation

has a lot going on and it's kind of exhausting to look at for hours at a time or you might make assets that look fine on their own but when put in context of being on screen all at once suddenly begin to fight for your attention because they're all moving and and and drawing your eye remember contrast remember how it applies to motion as well uh if your idle animation has as much motion in it as your run as your jump as your attack then the contrast between these different states is diminished and when an important state change occurs it can be harder to distinguish if they all have like the same amount of of motion going on you want to have uh breaks somewhere so that your eye can have a bit of a rest and assess what's what's going on uh the next thing to be careful of is thinking you know things

you don't know things i don't know things there's a lot of things i don't know and that's fine because things exist and we can look at them i'm saying use references even for things you think you understand use references browse the internet there's all kinds of videos on there uh you have a body i assume probably a phone with a camera film yourself act it out use props take note not just of the motion but of the sensation of moving where do your muscles tense and stretch what is uncomfortable to do what what position is comfortable to do is there a more natural feeling way of of doing whatever you're trying to do that this character could be moving that you hadn't considered you won't know these things all the time and that's fine because it's nice to have an excuse to act like an absolute

fool in a professional capacity i have been paid to swing a plastic sword around in my bedroom uh it was research uh my plastic sword broke last year i'm still sad about that i need to get another one uh it was an incredibly handy prop i recommend i recommend animators to have props plastic sword real sword maybe maybe maybe a little dangerous uh but really use references not only because you you'll have a better idea of what you're you know working on in the moment the thing that you're actually referencing but you'll be building a mental library to pull from which will be even more helpful on the next thing you animate or whenever you finally you know get to the thing that you need to to to animate that that doesn't have a direct analog that you can just look up you will at least have the collective knowledge of everything of all the references you've gathered of all the things that you've seen and the

ability to put them together in new and interesting ways to depict whatever action you're trying to get across lastly don't be too precious with your work i mean this generally but i also mean this very specifically in the sense that i sometimes see senior animators who they seem afraid to redraw frames like i see people draw a single frame of a character and then instead of drawing the second frame they take their one pose that they've made and are happy with and just kind of ship shift things around and rotate bits here and there maybe redraw things in piecemeal but more often than not it ends up looking like an obvious frankenstein variant of the one frame that they drew that they were happy with uh sometimes sometimes you can do that and it will be fine sometimes shifting stuff around is a completely valid way to save time and not redraw things that don't need redrawing

uh but being able to do that and do that well and more importantly being able to know when that is a you know thing worth doing is a skill it's a skill that you develop and and and the sense that you that you that you have uh but sometimes you just gotta draw six to twelve wholly unique frames uh and and you should i promise it'll be worth it you'll improve so much more quickly if you just embrace uh the opportunity to make new mistakes mistakes such as not keeping an object's mass and proportions consistent throughout the animation uh this is for real the last the last one uh this is one of those things uh where that old principle of animation that i that i uh glossed over is going to kind of come back to haunt us a little bit solid drawing uh that i mentioned and kind of dismissed earlier being good at drawing uh the best way to keep your animations

from turning to jelly as you work and and losing the mass and proportions is to practice and train your eyes so that you can spot when errors crop up uh hone your eyes to falcon sharpness also do construction and rough out your animation beforehand so you can catch those errors early construction in art refers to the creation of simplified forms to help in understanding the shape of more complex forms and the employment of those simplified forms as as a foundation upon which to build more complex forms so as an example you might draw a person's limbs as a bunch of cylinders to understand proportion without getting bogged in and trying to accurately capture the musculature of the arm in animation it can be similarly helpful to use simple shapes just to make sure you're keeping things consistently sized and

spaced without getting too hung up on the details you can see here uh animation of a bear boss for catacomb kids uh the rough version and the uh the final version uh i didn't personally do these but but uh they're a good example of of how you can just use blocked out shapes to to uh to test the motion before you lock yourself into any kind of rendering and spot spot issues early um here's an unfinished animation that i just started working on a few weeks ago i haven't made it much further than this yet but it's a death animation for one of the boss enemies in catacomb kids uh you know the one that was just on screen previously um i know for a fact like just looking at it i as i was working on it i was like oh there's a bunch of size inconsistencies and proportion issues and uh some stretches of time with you know i haven't figured out what the arms

are doing for the entire time yet but i'm not worried about it because the point of the rough is to get the motion right and the timing right and see where the problem areas are so that i can catch them and address them before they become larger issues if you jump directly into animating and rendering at the same time you're going to be pretty sore when you run into a big error in the very premise of the action that you're trying to convey you're going to be reluctant to address it because that would mean throwing away a lot of the work you've already done ask me how i know but you know it's like the old saying goes measure twice animate once so there are a few other miscellaneous things i want to touch on before i call it a talk um they aren't quite mistakes or principles or useful jargon or anything but they're just things i've come to understand and value over time that might be handy tools in your bucket of tools

first of all i think i should you know mention 2d skeletal animation i should talk about that for a bit a skeletal animation is when you have a character composed of predetermined bones and joints that are rotated on pivots rather than having each frame be drawn as a completely separate and unique uh image or frame um it's like how 3d animation works but you know one less dimension um and skeletal animation is great i i like it a lot it's i use it in my game catacomb kids for all the humanoid characters because it's important that they be able to change their visual appearance in accordance with their equipped armor and weapons and such i knew from the start that i wasn't going to have the wherewithal or patience or longevity of life to hand animate the dozens or hundreds of layers of equipment that would be needed to represent all the possible combinations in the game

and that's really one of the great benefits of skeletal animation you can reskin bones attach things to them uh on fly in engine to reflect aspects of the game that would be otherwise extremely difficult or time consuming to create if you were doing things in a more drawing oriented fashion skeletal animation is also kind of a requirement if you're going to be doing any kind of procedural animation which is animation that is done on the fly either partially or entirely by computation rather than being strictly handcrafted in order to be more reactive to the events in game and the environment uh skeletal animation also makes it easy to blend from one from one pose or animation to another smoothly without having jumps in in the uh in the animation from going from one frame to another uh changing pose uh but of course skeletal animation uh 2d skeletal animation is just as subject to all the previous principles of animation that i uh mentioned as frame by frame

animation is and it can be harder to implement some of these principles in a way that feels naturalistic and good if you're not careful because of the way skeletal animation often relies heavily on automated interpolation between key poses it kind of comes back to the idea that more frames does not equate to more quality in the same way i mentioned earlier that good in-betweens can't save bad keys the same is true in reverse strong keys don't automatically lead to good in-betweens or good timing so if you have if you have good good uh keys in a skeletal you know pose you still have to often really finesse the timing in the skeletal animation to get it to work well it's not a set it and forget it type thing um also in skeletal animation it can be tempting to make your keyframes and just save your time the effort of doing all the

in-betweens but that's again it's not a set of forget a type deal that's that's that's an easy way to to have your good art look bad it's an easy way to have your your as i mentioned earlier uh good style uh strong style uh can still be animated poorly and does disservice to everyone involved uh skeletal animation can also be a limiting factor to some degree in the kinds of actions that characters can pull off uh all of those aforementioned benefits come at a cost which is that if you want to make an animation that really pushes the action with lots of twists and flips like whatever the heck this guy is doing um it'll probably end up being much harder to pull off using skeletal animation than it would be to draw out the thing that you want to happen frame by frame and i also want to mention something that kind of irks me in particular that i see in a lot of 2d skeletal animation which is a lot of times the hips and shoulders are

neglected i see a lot of skeletal animation where the hips don't sway the shoulders don't shuffle as though those joints are pinned in place at their respective corners of an unrotating torso and these are two vitally important parts of the body for establishing the motion of the entire body and pose and they're central to the importance in creating forces that drive human action and i mean i kind of get it if you make the hips and shoulders rotate then you might have to then make a variance of the torso sprite to reflect the rotation and swap that out mid animation it's a little more work um but i don't know maybe it's worth it i admit i kind of have it easy in catacomb kids which is where this uh this green guy is from uh it's a grumble from my game uh i kind of have it easy because the sprites are so low resolution that i don't really have to worry about torso rotation uh you can't really tell it's just chunky pixels um so maybe i can get it away with a little

easier than than than others but if you're going to be making a game using skeletal animation it might just be worth considering keeping in mind uh what would need to happen to make you know to make the the the movement a little more naturalistic with the swaying of hips and shoulders and i know maxwell from scribblenauts is kind of a silly comparison because he's obviously very deliberately puppet-esque and cute and cartoony it's just this is just an example of what i'm talking about just to get the idea across i'm certain maxwell is a fine lad with upstanding character another thing to consider is this we're making games presumably we've got an engine that can handle certain aspects of motion and visual clarity for us image transformation scaling coloring rotation etc so you can save yourself some work and leverage those tools where you can squash and stretch as a visual effect is

relatively simple to implement in code and is generally useful enough effect that you might just want to say screw it i'm gonna just make a single frame for this thing and let the code do the rest uh if your character has a long scarf or hair that trails behind them maybe you don't need to animate it by hand maybe you can slap some physics on that element and have it flow and follow them in a way that both makes it more integrated into the rest of the game it also keeps you from having to animate a scarf effects like uh weapon trails and particle effects can uh can also be used and sort of offloaded to the to the uh engine side of things uh to better integrate them with the world as opposed to having uh having a lot of things baked into to animation so just something to keep in mind and while we're on the topic of engine stuff i also want to request

of the people who are making the tools and pipeline for games with traditionally animated art to consider supporting variable frame durations for animations the ability to have animations not play at a constant set speed of just one single frame rate the entire way through and to change the duration of specific frames mid animation is incredibly valuable for creating impactful timings i'm going to bring back the high knox just so you can see what a big difference it can make the one on the left is playing at a constant rate you can see the smear um that uh you know as i mentioned earlier you know smears are meant to clarify clarify motion but in this example you're not really supposed to notice it it's it's supposed to just be like very brief just there for an instance supposed to impart a feeling rather than being like a detail that anybody can

see um but the constant frame rate version that that frame has just as much uh time given to it and weight and importance as all the other frames of the animation and um yeah i don't like it so please please uh make your animators happy by by letting letting us uh have variable frame rates in our animations um that's it that's all i'm gonna say on that apparently that's all i'm going to say on anything um yeah that's me that's that's been my talk about animation and stuff i hope it has been helpful and useful um this is me forbid friday if you have any questions say what's up if you're gonna yell at me don't