Doran Meir - Stepped to Smooth

Heya folks,

haven't seen this here yet, so thought I'd put it up.

Israeli animator Doran Meir (who I believe is the inspiration behind Justin Barrett's tweenmachine) recently put up a tutorial on youtube on the benifits of working in stepped keys, and the million dollar question "How do you get to polished smooth from stepped?"

It's a very 2D approach. Personally I have found it useful to think (as Doran apparently does) in 2D terms, so first laying down my Key Story Telling Poses/Drawings (to have what the scene means), then my Extreme Edges of Motion Poses (to flesh out what's happening), then my Breakdown Poses (to describe how it's happening), and then my Inbetween Poses (to make sure the computer's not screwing up).



part 2
part 3
part 4
part 5a
part 5b

Scale Your Curves Up

Lately I've been working on some animation with a more realistic feel. It means there is a lot of time spent examining subtleties, and one thing you become painfully aware of is that when you're working on a piece that has lots of over lapping movement like a walk or run, its easy to loose sight of things. You can end up chasing little bumps and wiggles around in circles until you reach a point where you don't even know why they are there and what they mean. As your cycle gets nearer to completion it gets harder and harder to isolate the movement on particular parts of the rig.

An example might be in the movement on your characters shoulder, in my recent work I decided I wanted a sort of rolling action on the clavicle bone. I wanted it up higher as the arm swung forward and down lower as the arm swung back, creating a nice fluid circular motion. The problem was that with there already being animation on the pelvis twisting back and forth and bobbing up and down, four spine bones twisting this way and that and the arm swinging around it was really hard to see what that little clavicle bone was up to.

So here's the tip, I found the rotation axis for the clavicles up and down movement and then used the scale tool in the graph editor to ramp up the size of the curve and make the movement much bigger. Now the shoulder is making big over the top rotations when I play the cycle in the view port, with the animation still playing and the curve still scaled up I start to tweak it. With the movement so big I can see where the high point and low point in the movement is, and can see it I have the changes in direction happening as smoothly or abruptly as I want. Once I have the over the top movement happening as I want, I start to scale down the curve again with the animation still playing. Using the scale tool you can get the action to a size that looks just right, it may be lost is a see of overlapping action, but you can know that you have the movement you wanted in there and that it contributes to the believability of your movement. You can always scale up the curve again it you want to make further adjustments.

Measure Your Poses Energy

I'm obsessed with posing, so much so that sometimes I find it hard to muster enthusiasm for any animation or illustration I might have to do that doesn't have characters in it. But often I see student animation where the students focus on depicting the action has become a distraction from getting in some great poses.

I suspect that what happens is the student start to think in a linear fashion, or just follows one pose on from the next. Thinking things like, "well now the character needs to lean over in this direction, and the hand needs to move over there so it can touch that, then the foot needs to go there and so on." There is even probably thinking along the lines of, "I think that would look better if the hips twist this way, or I'll move that hand over there so it read clearer," and so on. But this is no substitute for taking time out to consider the overall pose at a specific moment.

Good planning with lots of lateral thinking, thumbnailing and acting it out is a good start. But then I think people get lost in the process and the poses worked out while planning can become a vague memory or feeling in the back of your mind while animating. So I thought I would try to introduce a new word into your animating vocabulary, its something that I find useful for staying true to my original vision for a scene.

Its simply ENERGY.

I like to boil down my vision for a moment in a scene to a description of the energy in the pose. Is it a high energy moment? Is it the highest or lowest energy point or do i need to hold something back for another point in the scene? Is it positive or negative energy? What direction is that energy moving? Is the energy directed out at someone or something, or focused inwards?

Then when I start to work on the pose (no matter what the medium) I start to hunt for that energy level in that moment. I will work a pose over and over (what was that Frank mentioned in the coments from the previous post about animation being "sculpture"), always thinking back to that simple, easy to remember idea about how the energy will work in that pose. I know that I will also have to make the pose work in with the movement too, but always try to remind myself that I can't sacrifice the energy level to achieve that.

With me there is almost always not enough energy in my first stab at a pose, my natural tendency is towards a stiffer less energetic pose than what I initially visualized. So the question, "How can I get more out of this pose?" is a constant companion for me. It might be worth thinking objectively about your work in general and how you want it to improve it, keeping the energy of your poses (and consequently animation) in mind may be a good way of nudging you work in that direction.

Anatomical posing tips

Hiya everybody, Alonso here.

Thought I'd mention a couple of anatomy facts that might help to make your poses a little more believable (if you're going for believable).



It's physically impossible to lift your arm higher then horizontal without also raising your shoulder. (This is because there's a little ridge on your upper arm bone (humerus) that connects to your shoulder blade at horizontal and pushes it up when you go higher)



There's a tendon in your pelvis that tilts your pelvis forwards when you try and reach behind you with your leg.



In a non cartoon human, the rib cage is a fairly solid structure so any bending and twisting is usually happening in the lower back, the neck, and in the shoulders. (so don't try to bend the rib cage in half, bend as far as you can with shoulders and stomach and neck.) So make your line of action, but then be aware of where the flexibility comes in when fleshing out the anatomy on top of the line of action. (The shoulders are super bendy, so they often disguise how stiff the rip cage is, because they cover the top part of it)

hope that's useful to some of you. :)

(sorry for my weak tablet drawings, but it's easier to jot out a sketch then find a suitable copywrite free image, so don't trust my crappy drawings stand up and test this stuff out on your own body, or do an image search for yoga and draw on top of it)


Aug '09 Oh, update! check the comments to see what my mom has to say :)

Ian Here - I like this post so much I'm adding it as a resource to the character posing section, thanks Alonso

Even Stuff

Here is my spin on why even stuff is bad when animating living things.

Firstly, what do I mean by even? Well its when things seem to have a pattern to them, sync up or move with symmetry. It might be two characters starting to move away on the same frame at the same speed, a blink where the eyes take the same amount of frames to close as they do to open, twinning in a pose is an evenness, or when a character makes several movements that all take about the same amount of frames. I use evenness to describe anything that gives the view a clue that the animation is made by a logical mind, instead of happening spontaneously.

You see the problem is that we are trying to convince people that they are seeing something alive, and as a general rule people (or any living organic thing ) don't move in such a predictable fashion.

When I have taught this in the past it is at about this stage that a\some students will challenge my generalisation. The will say that sometimes people do just happen to move in unison, or in patterns of timing, they might even be able to find an example of someone moving like that.

But there is a problem with relating these examples to your animation. When we see real people, we start from the point of believing they are real, they can move in many different ways and we still believe they are real. We don't stop believing in someone just because they do the robot dance. With animation the tables are turned, people start from the position of knowing it is not real, then we have to use our craft to get them to forget that. We want to eliminate any possible reminders that they are watching something that is made up made, and patterns, symmetry or evenness in your timing is a dead give away.

Of course unreal stuff can happen in your animation, but if so then we want to draw the audience in with a few tricks first so that they come along for the journey. We want our audience to believe in the threats and dilemmas that our characters face, even if the actual actions taking place are beyond the norm. Avoiding evenness id one way of drawing the audience into our made up world.

What we are doing is taking the subtle imperfections within the movement of living things and slightly exaggerating them in order to drive home to our audience that they are watching something alive. So take a moment to consider how organic the movement in your character animation is.

Frame Rate Shame Rate

I often see students asking about frame rates on blogs and forums. "What frame rate should I use when doing my student animation?" That sort of thing.

My new job is the first commercial one I've had working at 30 frames per second, and it has reminded me that it just doesn't really matter. You see your goal when learning to animate is not to remember how many frames it takes to do this or that, that's going to limit your vocabulary as an animator big time. And if it was really useful then the net would be full of long lists that dictate how many frames (at different frame rates) it takes to animate all the possible actions you might ever have to animate. Boy howdy, wouldn't that take the fun out of things, and make for some very predictable animation.

The skill you should be working on is your critical eye. Have a stab at the timing, knowing from the outset that you are unlikely nail it first go, then watch it and ask yourself how well it matches the idea you had in your mind. Critique it! Then set out to make the changes needed to get it working. Making animation is a process, not an event.

Why would you listen to the mathematician in your head, when timing is an opportunity to listen to the artist inside. Who cares about the frame rate, just make it good! Easy HA! ;)

Want to get better at animation?

Hey everyone, hope my ramblings are useful to some of you ;)


Here's something you can do to improve your skills, and you don't even have to animate ... give critiques. The easiest part of your animation abilities to build up is your eye.

You've heard the stories about animator's pulling in the janitor to look at their latest stuff? That's because every single human has been studying motion for their entire lives. So none of that "I'm not qualified to give critiques yet" excuses, our audience is everyone (not just other animators) as long as you have vision then you ARE qualified to critique. Critique's don't have to be long and elaborate, even quick little comments can unlock a puzzle.

The reason to give critiques is it will help you. The more you critique the stronger your eye gets. With a strong eye you'll be able to objectively see your own work, so when you're ready for a critique from someone else it'll be about making your shot sing instead of just fixing fundamentals to get it to acceptable. Practice at analyzing why things don't work and brainstorming ways to improve them directly will work just as well on your shot as someone else's. And the more critiques you give, the more will come back to you (like tag)

Eric has a great article on how to present your critiques in an easy to hear way. But what do you actually say? Ask yourself these questions:

1) Does it look like believable motion? (if you are new to critiquing, this may be as far as you can get, which is fine, just hearing "it doesn't feel believable" can help)

2) If it doesn't feel right, can you say why? (is it the way things move from pose to pose, is the problem isolated to only some body parts, are the arcs not clean? this is where the more critique's you give the more you improve this skill)

3) What is the animator's intention? (as an animator it's useful to know what people think you are trying to say, it can help you make decisions for better clarity)

4) What might help? (ways they could make their intent more clear, or their animation more natural. the longer you animate the more tricks you have you can share)

Warning, your eye will improve faster then your animation abilities will, which can drive you nuts. You may have upgraded your animation abilities to level 2 but now you have lazer precision level 8 eyes. But of course, with level 8 eyes you can analyze why your animation isn't working and hoist it up, and before you know it your a level 10 animator (of course your eyes will now be level 20 :P ... just keep swimming, just keep swimming)

So recap: 1) does it feel right? 2)why not? 3)what's the point? 4)things to try.

Touching Your Face

Something great about animation is that things you see in everyday life can inform you and make you better at what you do, Glen Keane talks about it here.

Recently I got sick right in the middle of an intense period at work, and because I couldn't take a break, the flu and following chest infection stayed with me for a couple of months. Yuck. This happens to me (and many hard working animators) a lot, working long hours for a sustained period of time is bound to run down your immune system. But there could be more to it, I saw a Doctor on TV talking about how people catch the Flu, he was saying that many worry about coughing or sneezing near them when the vast majority or infection happen when people touch their faces after physical contact with the virus from a surface. What tweaked the interest of the animator within me is that the doctor said the virus's were designed to take advantage of a "natural human behaviour".

The next day I decided to try and resist touching my face, and it's really hard! Plus the more tired I got, the harder it became (work long hours, get tired, touch and rub your face more, greater chance of getting sick!). Then I thought about how often I have my characters touch their face when I animate them. Hmmm maybe its something I could do a little more. What about you?

Quick! Hands on faces! by leslie adams

Find the Next Easy Bit

A crucial factor when animating is having a clear idea of what you want to achieve. That's why the nine old men placed such a strong emphasis on planning, they knew that when you can move forward decisively it empowers you to be the best animator you can be.

puzzle piece by Hjem

But no matter how much planning you do there are always going to be some parts of the scene that are clearer in your mind than others. When I'm animating I block out the poses that are clearest in my mind first, and when I'm polishing I sometimes work on the movements that I think I have the strongest feel for first. When you get these parts down, whats needed for the remaining parts often becomes clearer because you have context and contrast to consider. Its like a jigsaw, the closer you get to finishing the easier it becomes.

Use Percentages

Here is a little thing I saw years ago in an interview with Brian Singer when he made the first Xmen movie, I've applied it ever since and found it to be very useful. He was talking about what it was like directing a real professional veteran like Patric Stuart. Brian asked Patric to lift the emotion in a scene, he was surprised by the business like response from Patric, "how much more, 20%, 30% more?"

After Brian realised it was OK to talk to him about his craft in such a specific, calculating way he found it really easy to communicate what he wanted. Now I do the same thing whenever I am being directed or am directing.

That pose needs to lean in 20% more. . . . . . . . . . . . . That action should be about 40% slower. . . . . . . . . . . So you want that punch to have to hit 50% harder?. . . . . . . . . . . . So you want less emotion there, how much less? 10%? 20?

Clear, concise and to the point. You're happier because you know whats expected of you and your boss/teacher is happier because they feel its clear what is expected. Its so simple, but really helps me out in my day to day work.

Save a Key For Later

I'm starting a new label called Quick Tips where I can add simple little suggestions to help you with your animation. My first one is about animating lip sync.

When your animating mouth shapes it can be quite fiddly (depending on the rig) making the transition from open mouths shapes to closed mouth shapes. And seeing as how opening and closing our mouths is what we usually do over and over again when we talk, you can find yourself playing with the face controllers to reproduce the same action until your ready to pull your hair out.

Sometimes what I do is make two keys before the dialogue starts one with a generic open shape (a kind of "Aaah" shape) and and then one on the next frame with a generic closed shape (usually an "Eeee" shape because I find it most common). All the mouth shape controllers should be keys on these frames.

With the mouth shape controllers all selected I copy the key for the open mouth shape. If you made the closed mouth key second then the mouth should be closed through all of the dialogue. Scroll along scrubbing the sound until you find the first sound that will require a closed mouth, set a key, scroll along until you find the next spot for a closed mouth, set a key. Go back to the point between the two mouth shapes where the mouth will be open (assuming there is one) and paste the open mouth key. Work your way through the dialogue setting keys for the closed mouths, then going back and pasting the open mouths where need between them. Now you have a rough pass on the mouth animation that took very little time. You can playblast to check the timing.

Now you can go back and start fleshing it out with more specific shapes for the specific sounds and mor keys where needed.

Remember to make the sound with your own mouth, say the whole word. I can make an "EEE" sound with lots of different mouth shapes. You have to find the right one for this word, emotion and character.

Playblast your work lots and watch it over and over. This actually saves you time, trust me.


Hmmm, well I was hoping that quick tip would be a bit quicker. But there you have it :)