Alternative techniques were proposed, such as building a physical model and flying a motion controlled camera through. However, it is not possible to build a model big enough for a continuous five-minute ride. The three-dimensional part was a straight forward job for the computer graphics but the integration of the two mediums brought forth a few creative problems and some technical problems.
“Just to give you an idea of the magnitude of this project, it took a year-and-a-half from start to finish,” calculates Gibson. “It involved fifty eight computer animators and around forty cel animators.”
They produced initial storyboards with one or two sketches of the environments and the spaceships. These were developed into blueprints, models were built and then they produced an animatic. This was a hybrid of storyboard and animation, a kind of mid-step between storyboard and final choreography. When that was approved they created the camera motion through the environments with all the nuances and ups and downs that were needed to make it an interesting simulator ride. Finally they added the effects animation and the squash and stretch animation with the vehicles.
When the action was satisfactory they began to build the environments and colour them. They added hand painted texture to some of the environment, created the surface quality, the reflectivity, colours and so on. Individual frames were animated where required to get effects such as a sign going off and on. Motion tests were made of many of difficult segments.
A master was produced on 35mm four perf and then Vista Vision since this was released in 70mm for the theme park ride. They also produced a series of mattes for optical printing so the characters could be placed in the vehicles with some overlap; the mattes would provide a clean junction between where the cel animation stopped and the computer animation began.
“Finally there was a long, expensive, optical composite, which today we would do digitally but that was not available at the time,” disclosed Gibson. “Digital compositing is a tremendous boon to all forms of film making, not just animation, because of the fine control you have over the integration of images, and there is no generational loss. Opticals have been avoided classically in cel animation because of the generational loss, doing the work in-camera is difficult so I think the advent of digital compositing will really help.”
There were also some creative problems, mostly of a chicken and egg nature. This project was designed by computer animation directors who concentrated on the journey and gave less planning time to the cel animation. The computer animation was the long, slow part. After it was complete the cel animation was then produced.
“This is not the best way to tell a story but with the schedule we had it was the only way to do it,” admitted Gibson. “If we had unlimited funds and unlimited time it would have been great to produce some computer animation, give it to the cel animators, let them work with it, recommend some modifications, go back and forth, work with the voice track and so on. That was not possible on this project. Sometimes the cel animation was squeezed into the picture at the last minute. It would have been nice to equalise the priorities. However, it was a very good first project for all of us and everyone was pleased with the final result.”
Charles Gibson believes the barriers between the two media will continue to fall: “We will see more direct interaction between the cel animation directors and the computer graphics animators. Right now there are a lot of people in the middle and this makes the medium a bit inaccessible. As the software gets easier to use and more creative people move into the fold and fewer technical people are required this symbiosis is going to happen naturally and it will be a good thing.”
Another thing that will happen and is happening now is the technology is dramatically dropping in cost. Three years ago it would not have been possible to produce this project in the time. The amount of rendering produced in a very short time was possible because computers are now a lot less expensive and a lot faster. The benefit for most people is the bottom end capability keeps going up. “In a few years it will be possible to produce work of this type using very small computers operated by independent animators and small animation studios,” predicted Gibson. “The technology is here, it is being developed in personal computers and the accessibility will create new software that is easier to use and we will see everyone using this technology, not only a few specialised studios. The state of the art will also keep going up and we will see more ambitious projects on the high end.”
When questioned about prices Gibson said animation can be produced at many levels. Disney quality animation produced with cels costs between two and three-hundred-thousand dollars a minute. At that level cel animation and computer animation are very compatible in price. However, as the animation simplifies the cuts can be made in cel animation a lot more directly than they can in computer animation. “It is very difficult to streamline computer animation without compromising the look,” explained Gibson. “It is possible with design but it is not an easy problem. The cost is not all in machines at this point, roughly three-quarters of our costs are in labour. In the future our costs will migrate almost totally towards labour. This was not the case five to ten years ago, machines were then the major investment.”
page 1 | page 2
Printed in Animator Issue 30 (Spring 1993)