The initial concept for the dodos I was given was developed by Christine during the prototyping phase before I joined and looked like this:
One of the very first things I did on the project was to take this into Maya and begin blocking it out to see what parts of the concept worked in 3D and which did not. While doing this I was constantly checking my work against historical images of dodos, some remaining skeletons from museums, and images of chickens that had been genetically engineered to be featherless. One of the first things I noticed was that the legs would not work as drawn and needed to come off the sides of the bird rather than the bottom. On my next iteration of the blocking I found that the size of the beak and the placement of the head and legs made the dodo feel very front heavy, as though it would topple over instead of being able to stand. To compensate, I moved the head backwards and gave the neck something of an S shaped curve. From there I experimented with the general proportions of the model, the size and placement of the eyes, and adding/removing little bulges to make the silhouette more interesting.
Eventually (and after hearing that the beak made it appear to be a parrot rather than a dodo) I wound up with this:
The bird was much fatter, with an enlarged pouch at the base of a thinner neck (I wanted it to appear fragile and to make the head appear comically large in comparison). The beak was reworked drastically to emphasize the bulge at the tip and to slop upwards to connect with and be part of the skull rather than a nose-like protrusion. I'd also lowered the eyes and moved them laterally in hopes of making the bird appear to be less intelligent. Unfortunately, I didn't leave well enough along and things went wrong in the process of creating an actual mesh. The end result was this monstrosity, the abject horribleness of which may cause those foolish enough to look up on it to go blind. Fair warning.
Yeah, I don't know what I was thinking, either. I probably could have avoided things getting this bad by referring to the blocking a little more frequently instead of just using it as a jumping off point. Lesson learned. I wasn't by any means happy with what I had at this point, but it existed and a number of assets didn't and this marks the end of my work on this particular model as I turned 100% of my focus to other parts of the project.
Somewhere between the above travesty and the end of spring I was declared Art Lead on the project. At the end of the semester I took stock of what we'd accomplished, what we still needed to get done, and took my findings to the team and the professors. At which point we agreed that we needed some outside help in order to meet our goals in the time we had allotted. So over the summer I sent out some emails and in the fall the team met with a couple of undergraduate artists who had shown interest in the project. Sadly, while both of them showed a great deal of talent and enthusiasm for the project, one was let go as her fine arts background had not prepared her to work as a member of a team. Regrettable, but hey, the other artist gave me orthos, responded well to feedback, and gave me more orthos. I <3 ORTHOS.
Orthos become blocks.
Blocks become a mesh.
Mesh visits the local taxidermist.
Mesh receives a rig.
Dodo gets a new paintjob (eyes are procedural shaders baked to textures, everything else is Mudbox), some skin weights, and learns to walk.
I'm out of screenshots at this point, but from here I ran him through MotionBuilder to cut the animation up into takes. Which involved much wailing and gnashing of teeth because the .FBX format didn't much like parts of the rig. I got everything sorted out in the end with the help of my old buddy Charlie (who we shanghaied from another team in late September. Best. Kidnapping. Ever.). At which point we shoved the bird into Unity, slapped a proper shader on it, and started feeding the poor dumb bastards to giant Venus Fly Traps.
No comments:
Post a Comment