Background

Maya, oh Maya

This project was the result of a 2.5 month course taught by Enrica Lovaglio at Cal Poly, San Luis Obispo. While the primary learning objective of the course was to familiarize with 3D modeling, my term group consisting of myself, Ted Dumont and Kalen Goo chose to try our hand at keyframe animation as well.

Planning

Sketching, Storyboarding, and Struggling

Our collective vision was an animated music video, much like the classic Animusic videos, set in a post-apocalyptic city. Robot musicians would awaken one-by-one and play a song together one last time before shutting down for an indeterminate amount of time.

After agreeing on the project direction, the team compiled moodboards, sketched character models, and storyboarded the animated movie. One team member, Ted Dumont, supplied three original music clips which we narrowed down to the piece to be used for the music video.

Solution

Bringing It To Life

My main contributions to the project was some character modeling (laser robot dog), some environment modeling (street, trees), and optimizing render settings. The last of these was by far the most difficult but arguably the most important step in the process; without the proper render settings, the animation could take 300 hours or look nothing like it was intended to—or both. Despite the many challenges we faced, our team produced a 1-minute long, 3D animated music video as intended.

Lessons Learned

Give it Time

Being my first 3D modeling and animation project, I learned a considerable amount in regards to story creation, 3D modeling, nuances (read: frustrations) of Maya, keyframe animation, render setting optimization, collaborative 3D art, among many other things.

The most daunting task was rendering. Without access to the university’s lab computers that are intended for 3D applications, the 1-minute long, 24 frames per second video had to be rendered between the three team members’ personal computers. A first draft render optimization reduced the total render time to 300+ hours, but after further research and testing, I was able to optimize the settings to render in less than 100 hours.

Special thanks to Enrica Lovaglio for instruction, Ted Dumont for original music and environmental particle modeling, and Kalen Goo for character and environment modeling.