U.K.-based NaturalMotion developed endorphin 2.5 to allow animators to create high quality, finished sequences in record time.

For many years opinions have flowed on the subject of motion capture as a method for animating virtual characters. Whether they are for use in visual effects or computer game projects, the debate continues. There is the pragmatic, budget-minded camp that sees the savings (in both time and money) in reusing motions for multiple projects. On the other side is the creative perfectionist who requires such subtlety in his characters that the use of motion capture is an affront to his craft. Career animators can usually see both sides of the debate and almost always wind up hoping for an ideal middle ground. While it’s not the sledgehammer of finality this debate could really use, endorphin 2.5 is a leap forward for both sides.

Created by U.K.-based software developer NaturalMotion, endorphin is the first character animation software to combine dynamics with some fancy artificial intelligence, all in the service of creating believable, often realistic, character animation, practically at the push of a button. Animators need not fear for their jobs, though; the automatic results are nearly always stellar, but there’s also plenty of room for animators to have detailed control.

The paradigm endorphin follows is somewhere between an animator and a director. Using existing rigs from any of the industry standard apps, endorphin enables users to try out several solutions to animation problems, without going over budget. Game creators and vfx artists will find endorphin fits smoothly into their pipeline, usually requiring only a handful of extra steps for each character or sequence.

Familiarity with existing animation packages reduces the learning curve for those new to endorphin. All screenshots courtesy of Fred Galpern.

Animators familiar with industry standard apps, such as Maya or 3ds Max, will find themselves at home in endorphin right from the start. The main viewport and the tools for navigating are standard, including the mouse/keyboard combos and the keyboard shortcuts. The timeline also feels familiar, especially if you’ve used video editing apps such as After Effects. The event-based timeline system allows users to add what are essentially layers of animation to their characters. Forces, such as gravity, are represented in the viewport with three-dimensional icons that make placement and effect obvious and straightforward. Some of the other apps on the market would be wise to learn from the example set by endorphin’s simple icons. All too often developers create complex iconography when a simple arrow will convey meaning just fine.

“Dynamic Motion Synthesis” is the term endorphin uses for its slick AI driven character behaviors. Part of the system is a rich behavior set that allows characters to accomplish many simple and complex tasks. For example, users can direct a character to fall backward but extend the fall over an unusually long period so there is some dramatic tension to the sequence. On top of this users can layer additional behaviors. Animators may choose to have the character flail his arms or go completely stiff as he falls. The great thing about endorphin is how flexible it is. With many layers the animation might be expected to breakdown into unusable garbage, but it takes some sincere effort to force endorphin to produce a recognizably poor animation.