ANIMATION HISTORY 3D HISTORY CGI CAPABILITIES INFLUENCES EXPERIMENTATION
 
 
 
 
 
 

CGI CAPABILITIES

The capabilities of Computer Generated Imagery and animation are almost endless. I have briefly introduced the capability of 3D and cel animation so I will now begin to explore the capabilities of 3D in relation to my project. I will focus on the areas that are realistic for someone with a standard computer system and reasonable facilities available at University.

John Lasseter is one of the people responsible for the development of 3D animation as an entertainment. Before Lasseter showed interest in researching into 3D as entertainment, computer generated animation was only researched by science departments. He worked at Disney for 5 years but after Disney dismissed the idea of computer generated animation, Lasseter joined the special effects department at Lucas Films which later became Pixar. In 1984 animator John Lasseter teamed with technical computer scientist Ed Catmull at Lucas Films and created The adventures of Andre and Wally B, the first fully CG animated production (please see the ‘3D history’ section). The feedback from the film was overwhelming and it was clear that Lasseter had touched upon the 3D animation entertainment industry, and its potential was enormous.

In the publication titled ‘Principals of Traditional Animation Applied to 3D Computer Animation’ Lasseter documented the laws of animation adapted from Disney’s anthropometrically correct animation techniques and from Looney Tunes more loose exaggerated techniques. There are 11 animation laws Lasseter states,

1

Squash and Stretch

2

Timing and Motion

3

Anticipation

4

Staging (presenting an idea clearly)

5

Follow Through and Overlapping Action (interaction of actions)

6

Straight Ahead Action and Pose-to-Pose Action

7

Slow In and Out (spacing of in-between frames for subtle timing and movement)

8

Arcs (a visual path for natural movement)

9

Exaggeration

10

Secondary Action (an object resulting from another action)

11

Appeal

 

What can I do with 3D? -----------------------------------------------------------------------------------------

I am restricted to creating relatively low-poly models and minimizing the use of effects and plug-ins because of my systems capabilities, but without these restrictions anything is possible. Rendering will be a big factor too also consider (Pixar’s Toy Story took over 800,000 hours to render using their home developed rendering software Renderman). I will need to balance these factors when planning and designing my movie.

Character Animation -------------------------------------------------------------------------------------------

Chuck Jones famously said about Tex Avery, “I learned from him the most important truth about animation…Animation is the art of timing.”

The interaction of characters with their environment and with each other is essential to the fluidity and realism (to an extent) of the animation. Of course cartoons can tend to exaggerate actions that we may make but these motions are based on realism. Toy Story (1995 – Pixar Animation Studios) is an example of brilliant character animation and interaction. The idea for Toy Story (please see image below 1) was developed from an earlier Lasseter film Tin Toy where toys come to life and fear the child that plays with them.

 

 

 

The release of Toy Story marked the arrival of the biggest change in popular animation since Disney’s Snow White some sixty years earlier. The movie grossed over $390million worldwide. The number of fun, quirky and heroic characters, animated to an almost faultless level, is undoubtedly one of the main reasons for its success. They were instantly accepted as the children’s new favourite characters.

In 2001 after establishing themselves with Antz PDI/Dreamworks released Shrek. In the film character realism was achieved by modelling the characters with organic textures and dissecting their bodies into ‘real’ bone structures. This technique was time consuming but achieved remarkable results. It involved stripping the characters model down to a skeleton, then layering on the muscles and stretching skin over the top, much in the same way the dinosaurs were built in Jurassic Park.

Motion Capture -------------------------------------------------------------------------------------------------

The facilities at Brunel University (in the Bit Lab) would allow me to use motion capture techniques to animate characters. Motion capture also referred to as ‘mocap’ is a technique use to digitally record movements of the object linked up to it. The standard use for motion capture is to record the movements of humans. The technique can take a lot of time but the end result can be a fluent replication of the actor’s actions. However, problems may occur when the mesh of the 3D character goes out of line with the motion captured rig / biped.

Because motion capture only records the movements of the actor and not his/her physical appearance, his/her actions can be translated to any 3D model with the same body structure, such as an alien or a robot. This means a 3D character can take on the movement and mannerisms of the actor.

The procedure to record motion capture involves the actor wearing a leotard suit with a number of reflective markers attached at specific points. This can also be done without a suit but usually the results are not so accurate. Then cameras are set up to report the position of the markers several times per second. The captured information is then edited or tidied up in a 3D program such as 3D Studio Max.

One film that made great use of motion capture is The Lord of the Rings: The Two Towers (2002). It featured Gollum (please see below left 2) a CG (Computer Generated) character animated using the mocap technique. Only iRobot (2004) has managed a motion captured CG character at a similar level of fluidity and accuracy. However to gain the accuracy of Gollum’s movements, played by actor Andy Serkis (pease see below right 3), the team at WETA digital used both mocap and key-frame animation techniques. Serkis was quoted as saying “When I first came out to New Zealand for principal photography, the focus was on voicing Gollum. But Peter wanted Gollum to be the most interactive digital character ever made for a movie, with a combination of live action and digital animation.”

 

 

Gollum used up to 100 sliders, attached to each part of its 3D model and another 100 purely for the face, and each were animated individually frame by frame. The mocap was used on the movement of Gollum’s body whereas the key frame animation was used for the face and occasionally for the body in the less complicated movement scenes. Rendering Gollum and the other creatures in The Lord of the Rings trilogy required an array of 3,200 computers running 24 hours a day.

Comparing Off-the-shelf 3D software -------------------------------------------------------------------------

With the CG industry growing so rapidly and 3D software packages’ becoming more available at a lower cost the software market is larger than ever before. The use of 3D in advertising is so popular that it would be difficult to switch on the TV and not see a 3D model or effect (please see ‘intro’ section for a table of test results).

3D software was developed from 2D CAD packages and visualized what the designers created by using mathematical calculations. There are two major types of 3D packages, the solid modeller and the surface modeller. Solid modellers tend to be used in engineering as they produce accurate and detailed results, where as surface modellers are used for visual representation, this is what I am concerned with.

The surface modeller packages are primarily based around primitive objects, i.e. boxes, spheres, cones etc… Materials or textures can be applied to the objects and lighting and cameras can then be added. But as more is added to the scene, for example more lighting and a greater number of polygons for objects, the render time will increase. Rendering is a processor intensive part of 3D modelling and can take hours or days to render a full scene.

There are 3 main software packages that are affordable (well reasonably!) and popular. These are Discreet’s 3D Studio Max, Alias’ Maya and Soft Image XSI. Lightwave, Cinema 4D and Bryce are all equally capable packages but after researching on the web, the 3 I mentioned seem to be rated the highest. Although I have yet to hear anything too bad about Soft Image XSI, obtaining this package will be difficult as will Maya. 3D Studio Max is also available to buy on an educational licence from the University so that is why I have chosen to use that specific package.

 

 

I recently read an article online from a case 4 study that summarizes why 3DS Max is the software I decided to use,

I’ve always had a soft spot for 3DS Max. I’ve used other software but I keep coming back to what I believe is the best solution, for both new and experienced users. The user interface gives someone starting out easy access to the basic building blocks for modelling, texturing and animation and the more advanced tools are also readily available… 3DS Max 5 has an excellent toolset which provides great creative freedom. – Fred Weimer

After discussing my choice of software early on in the project with Fred Weimer (who has an extensive 3D background) I was soon convinced that the facilities and software provided by the University are best suited for a beginner animator who requires the tools to realize creativity.

 

-----------------------------------------------------------------------------------------------------------------------------------------

1 SOURCE: http://www.pixar.com

2 SOURCE: http://www.arwen-undomiel.com/character_bios/gollum.html

3 SOURCE: http://express.howstuffworks.com/gollum3.htm

4 SOURCE: http://www.manandmachine.co.uk/ mum/web/ezine/education/brunel.pdf