Deluge! p2: Tragic Kingdom

A few people already know that I’ve been working on a concept for a game; writing out scripts, doing concept art, etc. Well, during my last class (Intro to Games) I decided to take the opportunity to begin developing assets for it. First, some concept work for the first two characters (and the mascots of the game), Big Bad and Little Red:

Red is mostly a WiP, I haven’t quite finished painting her yet or doing everything I want with the background. I won’t go into the storyline yet, but there are many more characters to come based off their original incarnations, which were much darker “fairy tales” than people are probably used to. Who knows, maybe I’ll even inspire a new rating system, complete with hate mail from the AFA! One can dream.

As I said, more characters to come. This house is one of the assets I developed in the games course, created from modular components that were then put back together inside the game engine. Even with an extremely low polycount, it’s amazing what you can do with some decent texture work. Took hours and hours and hours of blending sometimes dozens of textures together to make them unique, but it was worth it. It’s rather pretty for a burned-down, decrepit pit of bad memories. This is Cinderella’s house, by the way. She might’ve ‘snapped’ just a little bit.

The following are just a couple of the additional many assets made over the course of the class for the scene, with more to come, including the house (not yet textured, not sure I’m going to actually need it yet) modeled as pristine and new:

Side note, I took that ornate gate into Second Life just out of curiosity using their new polygon-compatible software. Single digit prims even when blown up to a large size, can you believe? I might have to start taking that thing seriously.

Cinderella, Issun-Boshi, Chi Li and Prince Charming are on the chopping block next, though I may try and get some modeling done on one of these guys before I get too heavy into the concept art for the whole crew.

Advertisements

Base 18: Environment

Technically, the mesh is done, but as he was part of an assignment, I decided to give him an environment and animate a fly-through as well. It was a lot of work (all of this happened over 5 weeks), but it was worth it to get to pose the model and really flex the artist muscles.

The first thing I did, with the help of an environmental artist friend of mine, is spruce up and refine the design of an existing environmental concept. I was going for a sort of mystical spring at night, surrounded by trees. The backdrop was painted first, to reduce some of the workload, and placed on a flat surface. Two other planes were then created, one for the ground, and one for the water.

Next, the topology of the land is created with a combination of soft selection and manual dexterity.

Now, I’ve created some spheres and molded them into what, with texturing, will become runic rocks.

Next, the first tree is created from a cylinder, the branches extruded from select faces.

Some additional loops are created to define shapes, and give me more room to shape both the trunk and branches. As the tops of the trees will not be (generally) visible, they are left out.

The tree is then duplicated and further manipulated to go along with the original concept art painting.

The initial concept art is here. The couple in the water are sketched by me, the background was painted by the ever-talented Scott Xie, my friend and trusted environmental artist.

The video fly-through will be uploaded to the website and the link updated and posted here once that’s done. In the meantime, here’s a screenshot. As you can see, mushrooms were added to the trees, as well as water reeds, to give the environment more life.

Thanks for following along, folks!

Base 17: Skinning

This part is relatively easy, but time consuming, so I didn’t document too much of it. Basically, it takes part in two steps…two very loooooooooooong steps. First, use Maya’s capsules to get the general shape of the weights.

After that’s done for the various joints of the body, I then go in and manually fix it up with the weight painting tool. Not hard, but a little tedious. I had music to keep me company and keep it from getting boring.

Once all that’s done, we have the final rig.

Base 16: Rigging

Rigging is something I used to be afraid of, until I dove right in and discarded all apprehension. It’s actually rather fun, once you get the hang of it! The start is pretty straight-forward. For now, I just need a very basic rig. When I have a clearer idea of what I’ll need this model for, I can go in and tailor the rig to the project.

Once the profile-view joints have been placed, I move on to the arms. This is where it gets a little annoying. I’ve left my model in a relaxed T-pose to help maintain proper deformation during animation. It’s probably the better way to rig, but it does complicate some thing a bit. Moving joints around and trying to align them perfectly takes some time, and Snapping only shortens the workload so much.

The hands are done next, aligning them both with the mesh and the axis of the joints as well.

And that’s pretty much it! I mirror the arm and leg joints over to the other side, which is literally just a few button clicks. Afterward, I start building in the controls with which to manipulate the rig.

Special thanks to Digital Tutors for showing me some tips and tricks along the way!

Base 15: UV Mapping

While he’s not up for texturing as a base asset, I still want him UV mapped and ready to go. Each part of his body is isolated with a planar map, then connected where relevant in the UV window. While pelt mapping and other more automated tools are potential alternatives, I don’t feel they give me the control that comes with more manually dexterous methods when time allows.

Most if it is relatively straight-forward. The head tends to boggle some people though, so I’ve documented how I do it more carefully. First, I planar map each section separately: the face, the top of the head, the left and right side of the head/ears, the front of the neck, the left and right side of the neck.

Afterward, the UVs are stitched together so it’s one, flat piece.

Finally, the UVs are relaxed to provide even distribution of the UV Space. Notice the unavoidable seem is placed in the most inconspicuous area near the back top of the head, where hair is most likely to occupy and in the event of a bald style, where camera and shadow will obscure any potentially noticeable texture seams.

And here’s the final map.

Base 14: Completed Model + Count

And here’s the full model! I eventually played with him a bit more to tweak him for the final, recessed his forehead a bit and modeled some hair. Otherwise, here’s Kai!

Oh, almost forgot. I also took a look at the polycount and moved some things around to get it lower. What you see there in the upper right corner is the polycount of the subdivided, smoothed preview, so you’ll want to divide that by half to get the actual value: 1901 (roughly). Not too shabby at all. With proper normal mapping, the lower-poly version can look pretty close to that smoothed version without sacrificing on polycount. I generally use a subD proxy for general rendering instead, though.

Next, mapping and stuff.

Base 13: Ears

Ironic this should end up step 13. I’m not a big fan of modeling ears. They’re a necessary evil, but they’re also incredibly complex. I’ve seen people call this part “easy with practice”, but it’s yet to feel that way for me. But then I’ve seen those same people complain about hands, which I love, so…it works out.

Step one, draw the ear loops, followed by step 2, lovingly titled: delete stuff.

Next, I chose to isolate the ear and work on it without the head. I’d attach it again later, but this way I don’t get boggled down trying to work around a transparent mesh, clicking and dragging the wrong verts accidentally and what not. I start pulling out the connective flesh and shaping the lobe.

Now to shape the ear a bit more, I add a loop to give me more edges to work with, push pull and connect a few things and…

Now it’s really just a matter of molding it like clay, now that the map is drawn, in a sense…

Finally, it’s reattached to the head. I pull out the lobe just a bit more later on, and pull the upper ear away from the head more, but otherwise it’s done.