PDA

View Full Version : Planet of Karkaea; Kevin Edzenga's thesis WIP


trancor
01-19-2009, 08:28 AM
Senior Thesis - Planet of Karkaea
Created by Kevin Edzenga


. Hey, in this thread I'm going to be posting some of my work I'm doing on my senior thesis at the School of Visual Arts in New York City. To be done before April 6th 2009.

. The concept I'm going after is more of a tech demo than an actual story. This said, I'm not really going for a highly modeled, well animated thesis. I will be going more of a show of how technology can be used to produce some neat outcomes. I'm shootin' to become a Technical Director in the future. I'll also try to keep an up to date MEL script line count to help show the quantity of math and scripting used (I'll post back in a little bit with that count, a lot of scenes to sift through).


Here is a link to my current previs (Version 2) --
http://www.youtube.com/watch?v=C956bXFrVwk



The Main Script -
. "Crowded" :: Crowd Simulator, currently with logic intelligence
. I've written a crowd control simulator I'm calling "Crowded". Crowded is being used in multiple scenes of this thesis, being put to use in different ways.
. Crowded's animation works through combining custom animation on a character rig with mathematics to control exact turns or direction each character will choose to move in. For some of these crowd systems, no man made animation was used, that it was purely scripted.
. Right now, I wouldn't call it an artificial intelligence; I'd call it artificial logic. After thesis is handed in, the next step I take is to advanced this artificial logic system into a true Maya based artificial intelligence.



The Ideas Shown -
. Runner Crowd System
. Willow Tree Dynamics
. Lake Environment
. Fly Crowd System
. Lake Life Crowd System
. Icosaplankton Crowd System
. Baby Dragon Nymphs Hatching Crowd System
. Credits




Runner Crowd System --
. Crowded is pulling up custom animation cycles to animate the rig as the system deems. . Crowded is also allowing each of these little Runner creatures to have --
. A Field of Vision
. A Customizable number of frames that each character checks ahead to see if there are any collisions or obstacles so each Runner can react and move around or slow down for the obstacle
. Will move around other Runners in their path they are running in
. Will choose to randomly jump, run, walk, or to simply stop and just stand there
. Runners are able to run on a ground at any angle (still being worked on [see the previs for current state])
. Complete control over velocity (Direction movement and speed)
. Every Runner has unique color schemes; includes -
. Eye Color
. Skin Color
. Fur Color
. Each Runner also has unique --
. Speed of animations and running pace
. Offsets of the animations
. Seeds (A single seed number will cause the Runner to do the same thing every time the system is ran, change the number, and a totally different set of actions will occur)
(And I'm sure I'm leaving stuff out)

**The odd floating in the previs is from the jumping. The jumping math doesn't account for a slanted ground yet; in a few days that will be fixed.

http://thesis.metal-asylum.net/makingOf/thumbs/runnerGwahPoseTN.jpg (http://thesis.metal-asylum.net/makingOf/runnerGwahPose.jpg)



Willow Tree Dynamics --
. The weeping branch is an intricate rig; function ability includes -
. Distance between each leaf
. Distance Blend between Phi (1:1.68) and even spacing
. Leaf count adjusts based on leaf to leaf distance (15 max leaf count)
. Rotation between each leaf; from bottom leaf up
. Starting position down the branch
. Max Leaf Size

. I wrote a geometry to hair script some time ago. It takes any piece of geometry selected and moves and orients to the hair and is wire deformed by the hair.
. So in the end, you get a weeping willow tree with branches that will sway to forces, bump into each other, have dynamic weight.
. The tree itself is rigged as well, so the weeping branches have even more movement within them.

Willow tree with fewer leaves http://thesis.metal-asylum.net/makingOf/thumbs/willowTreeTN1.jpg (http://thesis.metal-asylum.net/makingOf/willowTreeBG1.jpg)
Willow tree with more leaves http://thesis.metal-asylum.net/makingOf/thumbs/willowTreeTN2.jpg (http://thesis.metal-asylum.net/makingOf/willowTreeBG2.jpg)



Lake Environment and Fly Crowd System --
. Using particles for a Fly Crowded system.
. Each Fly finds the closest Fly to itself and pivots around them, maintaining a distance between all of the flies in their swarming pattern.
**The Flies are hard to see in the previs. They are under the willow tree branches near the water.

http://thesis.metal-asylum.net/makingOf/thumbs/lakeEnvTestTN2.jpg (http://thesis.metal-asylum.net/makingOf/lakeEnvTest2.jpg)


Lake Life Crowd System --
. Any number of creatures can be added at any time. Crowded will find each creature and animate each separately. So far there are three creatures. Each uniquely animated through math.
. The system also takes into consideration depth of the lake. So certain creatures will spawn deeper than others.
. Each creature in this system as of so far has --
. Unique Seed per each character (giving unique animation to each character)
. Unique coloring and animation speed/movements based off the seed

http://thesis.metal-asylum.net/makingOf/thumbs/creatureFogTN.jpg (http://thesis.metal-asylum.net/makingOf/creatureFogV1.jpg)


Icosaplankton Crowd System --
. This system is the most math intense system of them all. Every Icosaplankton has 12 points enwhich other Icosaplankton can connect to. So through math and a logic system set up through a MEL involved rig, each Icosaplankton will animate until it picks a mate (Another icosaplankton to connect to). The creature will face the mate with the point it wants to connect with (finding the rotation value) and begin to move along that vector toward the mate (using linear math to find the destination for each Icosaplankton).
. Through math I've written a slow down system using a Sine Wave for the animation. When the Icosaplankton gets to a certain point offset from the other icosaplankton, they will connect and combine mass and velocity. This combined group will now be the target of other potential mates. Building chains of Icosaplankton through math.

Icosaplankton Organs Open http://thesis.metal-asylum.net/makingOf/thumbs/icosaplanktonOpenTN.jpg (http://thesis.metal-asylum.net/makingOf/icosaplanktonOpenBG.jpg)
Icosaplankton Organs Close http://thesis.metal-asylum.net/makingOf/thumbs/icosaplanktonClosedTN.jpg (http://thesis.metal-asylum.net/makingOf/icosaplanktonClosedBG.jpg)


Baby Dragon Nymphs Hatching Crowd System --
. Crowded used in a different way, using ncloth spheres, wrap deformed egg shapes, and the Nymphs which hatch out and will grow before the camera. Each egg collapses when the baby is hatched. Each Nymph has collision bodies connected to them so they can push past eggs that are yet to hatch or collapsed eggs in their way.
. The rig is unique that I can move the entire character back and forth on a curve, and have enough control that I can have the Nymph in baby form and have control over just that area of the body, and as it grows to full length can get controlled by more controllers on the rig through the process.
. Each Nymph can be a blend of two different diffuse maps.


http://thesis.metal-asylum.net/makingOf/thumbs/multiEggFogTN.jpg (http://thesis.metal-asylum.net/makingOf/multiEggFogBG.jpg)


Credits --
. The credits is where I use the Crowded system to animate 600+ flies to take on the shape of "Karakaea" and my name "Kevin Edzenga"
. Then, the flies move to the fly zapper in the background and the rest of the credits will show up.


- -- -- -- -

Feel free to ask about any aspect of my thesis, The Planet of Karkaea

desirefly
01-19-2009, 08:44 AM
nice work man.... cant wait to see this stuff in action

trancor
01-19-2009, 08:50 AM
Thanks Jordan

every day I'm a step closer to finishing

trancor
02-22-2009, 06:56 AM
Its been some time since I posted, but I've gotten a lot of work done in this time. I put an example of my egg/insect/hatching sequence that I'm working on now up on youTube.

http://www.youtube.com/watch?v=g0blur_CRNA

Well me what you think.


----

The egg itself is showing the different possible directions an insect can hatch out of the egg. Each egg uniquely colored. Also the ability to add custom crack animations through a manual control blend on the egg controller.

The insects walk/crawl is procedurally enhanced through setdriven keys controlled by distance nodes, add, multiply, and condition nodes. When the leg gets too far from where it should be, it pulls the leg back to the proper place. These controls can also be moved on top of this walk set up.

The hatch sequence is what I have up to this point in time. I project in the next 10 days or so I'll have pretty much all of my 3d maya side done. Then comes audio and after effecting.

thejudge07
02-25-2009, 04:42 AM
Hey Kevin, nice work man. I really like your ideas for your project and the animations look pretty solid. I hope it turns out that way you want it for your senior thesis due in April. Listen, I've been wanting to move to New York for a while, I applied to a lot of game companies and what not all over the place, and man..it's just really difficult. I want to know what insight you have for the job situation there in New York cause I want to move there ASAP. I really like the city and the jobs, it's just that I am not up to par with expectations. I graduated from college last december which was real recent, and the same day I graduated I flew over to NY looking for work only to come back empty handed back here in FL. Luckily I managed to find a full time job here, but not 3d related...the problem with that is, I barely have time to practice animation. So, let me know what your teachers advise you students to do as far as getting a job, cause I am really curious. You can email me if you want to Kevin. Thanks man, and keep up the good work!

trancor
02-26-2009, 07:15 AM
Thanks Jorge, I sent you a pm about the nyc situation.


5 scenes set up and ready to render the next time I get to the labs to use the renderfarm.
one of these scenes just needs another camera angle for a 6th scene. I should have some full comps to post by friday night.

CGTalk Moderation
02-26-2009, 07:15 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.