Ok, so now it's officially reading week for us in Edge Hill University, which is basically like a half term but with more studying and I can choose to go into the university if I so choose. As such, I'll need to explain what I'm going to do with this time to prove I actually did some studying.
My Hair-brained Scheme For Reading Week
For this, I'll be studying the book "The Animator's Survival Kit" by the late Richard Williams (of both Who Framed Roger Rabbit and Thief and The Cobbler fame):
Afterwards, I'll do an exercise or just discuss a part of the book itself to show that I have a good understanding of the material in question, so without further ado, let's get this party started.
Day 1: Walking (Page 102)
According to a piece of advice Richard Williams received from a guy named Ken Harris (and then subsequently gave to anyone reading this book), "A walk is the first thing to learn. Learn walks of all kinds, cause walks are about the toughest thing to do right". This is because every human being, whether big, small, tall or wide, walks in their own unique way. For example, someone with a lot of weight on them might lumber around a lot more than someone with not a lot of meat on their bones.
This is important because in it means that in animation (and media production in general really), the way a character walks can tell us quite a lot about who they are as people. To demonstrate what I mean, I'll be using LittleBigPlanet 3 for the PS4:
Here, we can see all of the characters in this video, who from left to right are Sackboy (the customisable hero), a Sackbot (which will come back later), a zombie (or just Sackboy in a zombie costume), a sleepy baby (which is just Sackboy in a baby costume), Avalon Centrifuge, Clive Handforth, Larry Da Vinchi, Eve Silva, Dr Herbert Higgnibotham and Victora Von Bathysphere, all have vastly different styles of walking from one another.
This can actually tell people who may not have played the game before (or to be more accurate, LittleBigPlanet 2 since everyone here has appeared in and/or debuted in that game), as Avalon's confident stride lets the audience know he's quite full of himself while Clive's more depressing shuffle tells us that he isn't confident in himself.
Meanwhile you have Sackboy himself, and he's an interesting case because his walk cycle is a little bit more generic due to the fact that he's supposed to be a blank slate that you customise throughout the game, so in service to this fact, they don't even have much characterisation of their own (in fact, a lot of characters call him "Sack Thing" to keep the player character gender neutral)
Meanwhile, the Sackbots (the wooden Sackboy robot and the Sackboy dressed as a zombie) are a lot more mechanical in terms of the way that they move since they're both technically robots. (the zombie's animation style is listed as "Zombot" in the game)
However, one interesting quirk of the game is that because they made all the levels with the in-game level editor (meaning they used things like And gates, switches and even Sackbots), all the other characters can also be unlocked as costumes for Sackboy to wear himself (or even herself if you make Sackboy into a Sackgirl). With this in mind, it means that the developers technically needed to make sure many of the animations would also work with Sackboy to make life easier for themselves and the community making their own levels, and how do they do that?:
Well as it turns out, the solution is to use the same rig across all the characters in the game (that use Sackboy's body type) to make animating them all easier since they would be able to switch animation styles to match the corresponding character. (I know technically they're all Sackbots but let's pretend that this is actually the Sackboy rig's skeleton)
What makes this interesting is that despite me stripping the skin off of all the characters, you can still tell who's who based solely on their walk cycle. This means that if I were to theoretically shuffle all the skeletons into different spots and ask you to identify who's meant to be who, you wouldn't have a very difficult time being able to tell due to the walk cycle alone.
Now that I've established how important a good walk is, let's do a little test meant for those that have never played or heard of the game before now. I'm going to play through the first proper level of the game's story mode and ask you to keep an eye on the character of Newton (he's the man with a lightbulb for a head and a purple coat), if you've never played the game before I want you to try and guess his role in the story based solely on the way he walks through the level. Ready? Let's do this:
For those of you that are rather observant, you may have noticed that despite his seemingly helpful attitude (what with him explaining the controls to any would be new player of the game and all), you can't help but shake the feeling that something is off about this guy due to the way he walks with his hands pressed together in front of him, almost as if he's got some sort of ulterior motive going on behind that goofy smile and friendly demeanour of his. (it doesn't help that he literally pulled Sackboy into Bunkum against his will)
This of course is to set up that he is actually the game's antagonist, as once he releases Bunkum's three titans, they take control of him and make him into a dangerous villain for Sackboy (with the eventual help of Bunkum's heroes, who are Oddsock, Toggle and Swoop) to save from himself.
I quite like this little bit of foreshadowing the developers did despite the rest of the game turning out to be a bit of a rushed mess at launch.
Day 2: Changing Shapes/Easing In and Out (Pages 50 and 51)
Today, we'll be looking into how animators use changes in shape and even speed to convey different actions.
Richard Williams says that a master animator named Eric Larson would tell the young Disney animators he instructed that "what animation has to have is a change of shape" (WILLIAMS R, 2009)
The main example Mr. Williams cites is turning a closed hand into a point:
The gist of his example is that easing in and out of key poses (or "extremes" as he calls them here) will make the hand change its shape in different ways depending on how you go about implementing your easing.
If you were to ease out of the key pose, the hand will spend around 3 frames closer to the initial state it started in, meaning the hand will appear to reach the end state faster. Meanwhile when you ease out, it will spend those same 3 frames closer to the end point, making the trek to the end state feel slightly slower because you took what would have been frames 2 and 3 and changed them to be frames 3 and 4:
Here's easing out:
And here's easing in:
If you wanted to do both in the same animation, you would get what Mr. Williams refers to as a "relaxed, slower action", although you'd end up needing more frames (making the total drawing count go from 5 to around 9):
As you can see, each of the three types of easing can change the way your animation flows, even if subtly. This means that if you learn to understand the use cases for each type, you can make anything move in as convincing of a manner as possible.
I personally found demonstrating the three types of easing via OpenToonZ to be quite fun because I was able to get the hand drawn quickly and use the onion skinning tools to gauge where the hand should be on the next frame, plus drawing out the two extremes first turned out to be quite handy, as I was able to keep those two frames the same across all the animations while every other frame was different. It also helped that once I was done with one, I could delete the frames I didn't need anymore, do the other easing tasks and save those separately from the original. Another added bonus was that because I did this digitally rather than with paper and pencil, I didn't need to keep track of dozens of separate frames all at once while fearing that I might lose one in the process of making this. All I needed to worry about was the one viewport, the timeline and that was it. (All the frames are stored digitally so I also can't accidentally lose them)
This concludes today's study session.
Day 3: Turning Heads (Page 87)
Today, I'll be animating a head turn with guidance from Mr Williams himself (or at least his words), but before I do that I should explain some of the minutiae surrounding this that the book itself details:
Basically, Mr Williams explains that if you were to do straight in-betweens for a character turning their head, it will look like the features just slide across the face like so:
Of course, something like this would be fine if the objective were to create a turn around of the character where they remain absolutely still or even if the camera was the one doing the movement, but unfortunately for us, that's not actually the goal here.
No, for us the goal is to make the character turn their head of their own volition without the help of the camera, so what we'll need to do is to make the character's head movements appear more natural than my deliberately bad example from above.
We can do this by displacing the in-betweens ever so slightly to give the illusion that the character is bowing his head down while he's turning towards us and even adding a blinking motion to the whole thing to give the character a little bit more life while doing the movements as you'll see below:
I know my work here looks quite rough but I was trying to get it done fast for the sake of this blog post, a proper animator would make sure everything looked nice and consistent while the motion was in play.
To be honest, today was more of an excuse for me to go ahead and use OpenToonZ to turn Super Sheep's head around. (of course, the rest of the week, except Wednesday, was and will probably continue to be an excuse for me to try out various animation exercises, but this day in particular is a blatant excuse to do this)
コメント