Doc:Tutorials/Animation/BSoD/Principles of Animation/Principles/Straight Ahead Action and Pose to Pose

From BlenderWiki

[[{{{2}}}|      ]]>

[[{{{3}}}|      ]]>

Blender Summer of Documentation: Contents | Manual | Blender Version 2.41

[edit] Introduction

[edit] What

Straight Ahead and Pose to Pose are work methods for animators. They are possible answers for the “how do I animate?” question.

[edit] Straight Ahead

This is the oldest one. It means animating sequentially, creating pose after pose, frame after frame, "straight ahead", from the first to the last in a sequence.

Its main characteristic is that while animating one has considerable creative freedom to choose how and where each pose will be.

[edit] Pose to Pose

Following this workflow, the artist starts blocking out (planning, sketching) a sequence, by defining the key poses in it and roughly estimating in which frame each of them should be. These poses, which are known as extremes, are created first.

The remaining, "transition" ones -- called inbetweens -- can be done automatically by software interpolation, but for acceptable results, in particular to apply the principles we're studying and to remove the mechanical look of computer animation, further work is needed.

[edit] Why

[edit] Straight Ahead

Good points:

• very open for creativity during the keyframing work;
• can result in more fluid and natural looking animation;
• done for all frames in a sequence it totally eliminates the use of computer interpolation;
• a talented artist in a good day can achieve very spontaneous and elaborate results;
• specially good for fast, wild actions;
• only after finishing the animator will know how the scene ended up.

Bad points:

• it's easy to come to dead ends, where work may have to be discarded and redone;
• hard to make the character respect "marks": be somewhere or grab something at a definite frame;
• hard to create strong poses, well staged, solid and appealing;
• the resulting set of keyframes will probably be complex and disorganized, hard to work with;
• only after finishing the animator will know how the scene ended up.

Some might also add: more work while animating.

[edit] Pose to Pose

Good points:

• more control over the results, easy to respect marks;
• less room for pitfalls;
• more work in less time;
• the created extremes (plus soundtrack) can be used as animatic for critique and corrections;
• with planning, there's a better chance to come up with elaborate poses and moves;
• done every N frames or so, results in a clean layout of keyframes, easy to work with;
• randomness is played down.

Bad points:

• easy to result in “robotic”, dull animation;
• less room for creativity while creating and keyframing poses;
• randomness is played down.

Some might also add: requires proper planning, sketches, etc.

[edit] How

[edit] Preparations

Animators must have a clear understanding of the scene before starting, of course: where characters should begin and end, what happens in between and how they react to it, how long should the sequence take, etc. Besides the written story, storyboards and animatics (we also recommend this one) can be great tools for reference and guidance.

A simple “scene start-up kit” for an animator would include data files with the models (rigged characters, scenery, objects) and a soundtrack, at least with the dialogues, if any.

During the work the animator will playback (parts of) the animated sequence to check how it's coming along and to correct any problem.

[edit] Animating: Straight Ahead

1. create initial pose at the starting frame;
2. insert keyframes for the created pose;
3. advance frame by one;
4. update pose;
5. insert keyframes for the updated pose;
6. go back to step #3, continue animating “straight ahead” until the last pose has been keyframed.

[edit] Animating: Pose to Pose

Ideally, the number of frames between each extreme should be similar, it's not a good idea to have two extremes with many frames in between. Anyway, experience will tell you what works better and, since it's easy to shuffle keyframes later and add or remove poses, a beginner should not be stopped by the choice of extremes or the frame each of them should be put.

Working in layers

First pass

Create and keyframe the main poses (extremes) in their corresponding frames.

Second pass

Add anticipations,for example.

Third and further passes

Add other details, fight tweak and shuffle poses around if needed, edit animation curves, fine tune. Considerable part of the latter passes will be to try to make the animation seem more natural (less mechanical).

[edit] Animating: Pose to Pose + Straight Ahead

Both have their advantages and problems. The best approach, then, can be to mix them:

Hibrid method

1. start with Pose to Pose;
2. in one of next passes, maybe only for certain key parts that don't look good with software interpolation, return and fill in between the already keyframed poses, animating straight ahead, substituting the interpolated data by new keyframes.

Good points:

• unites the best of both worlds;
• avoids the main problems with each method when used alone;
• can give a wealthy mix between tight control and creative freedom.
• Pose to Pose is a good overall method, but Straight Ahead is better for faster actions (so an animator may end up doing frame by frame on such parts);
• Straight Ahead is not a good idea for mechanical motion and anything that can be interpolated well with animation curves, for which Pose to Pose works very well.

Bad points:

Actually, nothing not present already in the two methods:

• requires planning for the Pose to Pose part;
• it's probably slower than using Pose to Pose alone;
• gives a little more chance for pitfalls in the Straight Ahead parts, though probably in much smaller and easier to fix steps than when using Straight Ahead alone.

[edit] Notes

[edit] First Impressions

An advice from the book Illusion of Life: do not animate the scenes in the order they will appear in the film. The opening scene should give the best impression possible to the audience, thus it's better to animate it after getting more experience with the characters, gained by starting with one or more easier intermediary scenes.

[edit] Motion Capture

Using Straight Ahead Action is like sampling our idealized notion of movement by hand: motion in real life is continuous in space and time and happens in chronological order. With Straight Ahead we're trying to encode analog (continuous) movement into digital (discrete: in frames) one. Pose to Pose, on the other hand, is achieved by looking at real life movement as a collection of well define bits called actions, which can be decomposed in a few more representative poses.

Seriously, though, we also recommend that you learn about another method to produce animations: motion capturing, which does use sampling of real life movement.

More: Motion Capture, Sampling.

[edit] Good Tutorials

There are some very good and thus well known tutorials about animation methods and workflow at Keith Lango Animation.

[edit] Definitions and Nomenclature

The Toonopedia website has a glossary about animation.

Summer of documentation 2006 -- Willian 07:20, 5 July 2006 (CEST)