|
 |
|
|
| The Ongoing Trilobite Reconstruction Project |
|
Page |
1 |
|
2 |
|
3 |
|
4 |
|
|
 |
January 4, 2005 |
|
|
| I've been working on the best way to simulate water. I've read quite a few tutorials and now have a pretty good idea of the best way to go about it. Primarily, it's a simple plane, made transparent and reflective, distorted and deformed using various fractal noise functions. The color comes primarily from the colors you use to define the environment, in this case, a simple gradient.
I'm very happy with what I have so far, but I think a few more adjustments will make it quite perfectly realistic.
|
|
|
| Water moving. Click the above image for a six-second movie of just water.
Quicktime, Sorenson 3, approximately 300 K.
|
|
|
 |
|
January 9, 2005 |
|
|
| An incremental advance in the water texture. This was rendered without any antialiasing, so it's a little crude, and the textures, etc. still need adjusting, but it's an experiment in adding a sky to the water render. As a proof-of-concept, I think it works. What I did was to render four separate passes:
1.)Plain water.
2.)Plain sky.
3.)A masking layer, done by applying a flat, white color to the water, and a black fog. This allowed me to blend between the clouds and the water in a controlled way.
4.) A displacement layer, done by having the water surface reflect a red and green gradient. This let me ripple the sky image in a way that allowed it to seem like it was being seen through water.
|
|
|
| Water composite test, click the above image for video.
Sorenson 3 Quicktime, approximately 2MB
|
|
|
|
 |
|
|
|
This is the masking, or matte, layer. Because of the way After Effects handles masks, I had to invert it (make it a negative). |
|
|
 |
|
|
|
This is an example of the images I generated to displace the sky image. By rendering in two colors, I was able to set a different horizontal and vertical displacement based on the amount of red and green in a given pixel. |
|
|
 |
January 15, 2005 |
|
|
| Another revision of the water. This time, I used a different type of noise for the ripples and added a layer of particles to add the sort of little floating thingamajigs that give the water a sense of volume. I'll be revising those as well, since it will take a while to get the right balance of "almost-thereness" for the particles.
Click on the image at left for a six-second Quicktime movie. Sorenson compression, about 3.5 MB
|
|
|
|
 |
|
March 26, 2005
|
|
|
|
I've spent the last few weeks working on an animatic for the animation. If you're not familiar with the terminology, an animatic is sort of an animated storyboard, designed to allow the filmmaker to see how timing, pacing, and story elements work out in advance of doing the hard labor of animation setup and rendering.
I've decided that the opening scenes of my little film will be a "pan-and-scan" (think Ken Burns) of a painting of the Ordovician ocean. This will let me introduce the creature in an interesting way while still reserving the animated reconstruction for later, as a payoff. It will also allow me to do a painting of a bunch of Ordovician life.
And, yes, that's me reading the voiceover.
|
|
|
| Click image to download animatic. Quicktime, Sorenson 3, Approximately 550 K |
|
 |
|
May 3, 2005
|
|
|
| I've spent the last few weeks figuring out ways to procedurally animate the trilobite. What that means is setting up relationships between elements such that I retain control over what the thing is doing but don't have to hand-animate every leg and flexion.. In this case, I set up one control that governs the speed of the leg-kicking, and another set of controls that determine how much each joint of the legs will flex, affecting all of the legs at the same time. This lets me make sure that the legs won't accidentally pass through one another while waving. I've also been watching a short video of Triops cancriformis that I ordered from a wonderful scientific video house. The video dates from 1965, and is shot in high detail and shows the legs' motion in a variety of circumstances and in slow-motion. If you're into this sort of stuff, it's well worth ordering. I just wish it had been on DVD.
Anyhow, now that the legs have been worked out all I need to do is set up the coloring (I have a great couple of photographs of a Glyptonotus antarcticus, the marine isopod, for reference) and I'll be ready to animate!
|
|
|
| To see a three-second rendered test of the leg motion, click on the image.
Quicktime format, Sorenson 3 codec, 675K
|
|
 |
May 6, 2005
|
|
|
| I've spent the last couple of days perfecting the rig on the bug and working on the coloring. The rig is pretty good, although probably more complex than it has to be. The little star-shaped things in the scene are "nulls", basically just points in space with no dimension, used to control how elements that are visible move around. I've been running little animation tests, trying to make sure that everything works as expected before I go ahead and try to animate. It's awful when you find things acting unpredictably just when you thought it was all set to go. Surprises are not welcome in animation, because they're rarely good surprises. To check out one of the tests, click the image to the right. |
|
|
| Rig testing animation, about 1.4 MB, very compressed and artifacted.
QT format, Sorenson 3.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|