Article Focus: Peter Wiggins takes a thorough look at the new Motion from Apple. Never before has there been such interest and anticipation over a piece of software and Peter Wiggins hopes that this article gives an insight into the creative capabilities, workflow and SPEED of Motion.
Behaviors are the main feature that make Motion standout from the rest of the TV graphics applications. Traditionally, to animate any object, most other software packages resort to the user having to enter positions of an object in time, or keyframes. Now, this might be quick enough to set up when moving an object in a straight line, but it can become a nightmare when an object has a complex motion path or has to interact with other elements.
Motion takes a completely different approach, although for spot on accuracy or timing, you can use the keyframe editor.
There are four types of behaviors in Motion, all found under the tab in the library.
Basic Motion contains the most frequently used, such as Throw/Spin/Grow/Shrink etc.
Parameters are slightly different as they act on selectable values in other behaviors. An example would be to apply Oscillate to a Spin behavior. Instead of the object just spinning in one direction, it would slow down and then rotate the other way before slowing down again and spinning in the original direction; repeating the process over again.
Simulation These are really cool one click behaviors that actually perform some complex moves on text or objects. As the name suggests, these replicate some real world forces like Gravity, Drag and Collisions. Duplicating some of these would be almost impossible using only keyframes.
Text Due to the essential use of text in motion graphics, there are two categories for behaviors especially and only for text (Bearing that any of the above can be used as well) Text Animation contains the expected basic text moving functions like Crawl & Scroll. Text Sequence contains a list of ready made advanced text animations all of which come apparent as you get a preview thumbnail movie in the library as you click through each interesting move. You've probably guessed this is how I made the section headings for the article.
This last set alone will save you many hours of keyframing, and remember that behaviors can be stacked on top of each other on clips, text, objects or layers.
Let's take a look at a fairly simple example, again it's a lot easier to show than describe!
I have dropped a looped animation of a rotating dollar on the canvas.
Here I've just applied a Throw and the dollar starts travelling across the screen from left to right. The direction is set by the direction of the arrow in the dashboard and the velocity is set by clicking and dragging to adjust the size of the arrow. One tip here, if you hold the `alt` key whilst clicking, it will only alter the speed and not the direction so the object so can stay on a predefined route. If you run out adjustment with the arrow, the inspector takes larger values.
By stacking the Gravity simulation on top of the throw, the dollar now falls off the bottom of the picture, accelerating as it goes.
Applying Edge Collision simulation causes the dollar to bounce off the bottom of the frame in reducing parabolas. You can also see that as all the edges are enabled by default, the dollar also bounces off the right hand side of the frame.
One more simulation and that's Drag.
Now instead of bouncing off the edge, Drag has caused the dollar to slow down and come to rest. This motion path has taken 30 seconds to make, something that would be almost impossible to duplicate accurately using keyframing.
Complex actions and interactions can be built and maybe more importantly changed in seconds. Take a look at the excellent Pong simulation that Jayson Steckler at Apple built.
OK, but how is a bouncing dollar or Pong made without keyframes going to help in everyday motion graphics work?
Let's take a clip of a high jumper and over that draw a white square.
Then duplicate the square layer a few times and apply a Vortex simulation to the original box. This causes the other squares to rotate around the original. I've toggled off the background high jumper just for the moment so you can also see I've adjusted the transparency of the boxes so they overlap to give a bit of depth.
Now, toggle off the original box in the layers tab.
We are building an animating mask that I'm going to use to cut out the high jumper, so let's place the boxes onto of the video.
I've positioned the boxes over the area I wish to cut out. Note the picture has moved as well as this is also under the Vortex simulation. This could easily be deselected or separated by making another layer, however for this example it actually adds to the movement.
Toggle the blend setting to stencil alpha and the rotating boxes become a key signal to cut a hole in the video.
I've also added a Throw to move the composite across the screen,
and a Fade In/Out. It's important to remember that I'm applying all these behaviors whilst the clip is playing and looping in real time.
Now a web page is hardly the best medium for showing moving graphics, but this sequence gives you an idea
Looks fantastic for five minutes work, but it's against black, hardly an inspiring backdrop if you were making the programme titles for example. Remember that using stencil alpha has generated a key signal equivalent to that of the rotating boxes.
To composite it, let's pick a background, a head on replay of some athletes running towards camera.
Now, we are going to do another stencil alpha, but this time using an emitter as a key signal, this is Light Valve.
With a bit of moving and re-timing of layers, we now have our final composite.
Complex animation, all done without a single keyframe! The power and sheer speed of using behaviors is truly amazing.