Sprite Fright: Animation Tips Learned The Hard Way

Note: the following piece contains major spoilers.

Sprite Fright is the new open movie, now in production. In this article, Rik explains what he’s discovered while working as the project’s lead animator.

Planning, planning, planning

Diving straight into a shot can be overwhelming. Spontaneously generating ideas while pulling bones around is hard To help, I’d suggest addressing the following points before you start.

• Check the shot’s contents (“What is going on?”). Also, determine the “nature of the shot.” Is it action- based? Performance-based? What emotions are at play?
• Take notes during the Director's kick-off meeting (in which he explains his expectations for the shot.)
• Thumbnailing. Sketch the most important moments. Write down important dialogue (the voice-acting).
• Shoot reference material (e.g. If the character has a particular walk, film yourself performing that walk).
• Generally, the more information you gather, the easier the blocking phase will become. You’ll have lots of beats fleshed out before you begin.
• Determine your deadline, and plan accordingly.

Don’t be afraid to share work with a supervisor, the director, or your peers

• At around a quarter of the given time, share your work with your supervisor. (I'd say earlier than a quarter)
• Avoid doing unnecessary work: keep it simple and to the point.
• Even though your work might be rough, it's good to get early directions/changes so you can iterate more easily.
• Work from rough to detailed. Don't put in unnecessary animation in the beginning (blocking) phase because it may distract from your fundamental choices.
• However, sometimes you'll need to sell a new idea so consider giving it some extra love. With a little more time and flair, you’ll be able to sell your idea to the director.

Animate drunk, polish sober

Once you have a clear plan, open Blender. Then what?

When animating I like this motto: “Animate drunk, polish sober.” Well, what does that even mean? Every new shot is essentially a blank piece of paper: characters in T-poses staring into infinite space. How do you turn this into something believable and entertaining? I usually approach this problem as though I’m playing with lego.

If you look at the actual Lego Movie for example, they are basically doing exactly that and getting away with it. The animation is entertaining and believable within that world, despite the fact that it is oversimplified. Somehow with very simplified motion we can already read actions, acting beats, even emotions.

▶️ This video is only available for Subscribers.

It tends to strip away the unnecessary details, reducing an acting beat to its essence. Therefore I tend to only animate the body bone from the start exactly as you would when playing with a doll. My first objective isn’t to worry about the face or fingers; rather, I try to get the story beats right.

It's like determining the rhythm of the shot. Where does the character change her attitude? Where should I accentuate a line? Those beats can be easily set by simply animating the character like a rigid puppet. And you immediately feel if the motion is syncing up with the audio If that's the case, and you had it in mind, you can do the actual posing on these beats. For me personally I'd like to do this (mostly) in 'Stepped' keys, or 'Constant' interpolation (for those of you who speak ‘Blender’). This method gives me clear control over every movement Nothing moves by accident, so everything is a deliberate choice. This phase is the part of animation where we have full freedom to explore different ideas. Keeping it simple makes iteration easy, and motivates me to try new things without being limited by the complications of the entire character rigs. Once a certain idea is approved by the director, we dive into the details: making the performance believable, appealing and entertaining to the audience.

Train your 'animation' eyes

When watching animation from an analytical viewpoint, sometimes you might be able to recognize that something is 'off, but you can't quite put your finger on it. In order to become better at animation it is important to train your eyes for 'good' and 'bad' animation. What makes animation appealing, or why does it appear to be broken/unbelievable? You might know about the twelve principles of animation, created by Disney's nine old men back in the day. That's a good starting point to learn to look at animation. Looking at a piece of animation frame by frame really helps to analyse what the animator did (or did not) do to achieve appeal, good timing or a fun character performance.

By looking at an animation shot more from an analytic viewpoint, you will be able to train your eyes to look for possible improvements or simply to admire a well-executed shot. By looking at how these principles are applied on different shots you can learn how to apply them yourself and add them to your toolkit . When I do a polishing pass, and really go into details like fingers, facial features etc, a lot of the time I refer back to these lessons learned from my own experience, tricks I picked up along the way.

Set up your shot properly before animating (Yes, this seems contradict point three, but it doesn’t)

This is especially applicable to more technical shots involving a bunch of constraints, parent-switching, moving cameras etc. It can be a trap when you just start doodling (as I was preaching in point three ) and later on figure out that you have to adjust the timing, or change cameras, you'd better start over again. Having a solid plan for your constraints and animating the right controls can really save you a lot of time down the line, even though it might take a few hours to set up from the start. Think ahead: when do you need to switch from FK to IK? When does the character get pulled away by the sea monster so I need to switch parenting? Now, once this is set up well, then you can go to step 3 and let your creative ideas flow in abundance

Rik has helped document Sprite Fright throughout the film's production. You can find further articles on what he's learned here and here.