Creature of the week: Quanazillicus

The Quanazillicus prefers to go by Quana, for short.

Spore Creature Creator is here

I've been surprised the last couple of days just how excited I was about the Spore Creature Creator (released today). So I downloaded it the minute I got home from work, and I've been playing with it for hours. As expected, it's tons of fun. A stripped down version is available for free (you should definitely check it out), and you can get the full version, complete with hundreds of creature parts, for $9.99.

Unfortunately, the full version for Mac won't be available until tomorrow. Not sure why the delay, but I suppose I can handle it. If you're interested in how it works, there are thousands of videos and tutorials on YouTube (the creator allows you to post videos of your creations).

I'll avoid elaborating further (you can read a good overview here) and simply tell you again to go download at least the free version. All your creations can be made available to populate other players' universes when the full game comes out in September. For inspiration, check out a few of the creatures I've made today.

3D without (extra) headgear

Not much on the game development front to report today. I've been doing some modeling in preparation for my next project (or at least a pilot test of some concepts). Progress is slow so far. Even with a decent shortcut for character modeling, there's still a lot to learn. The hardest part has been keeping the model simple and clean so it doesn't have too many polygons and will deform properly once it gets into the game. Character design is quite tricky.

Regarding the title of this post, I saw an article this morning about a new television technology that can simulate 3D video without goggles. It basically sends two signals to the screen for every frame -- one contains the image itself, and the other contains a grayscale version of the image, which conveys the depth information. The grayscale image is used by a thin layer of...something...laid over the screen that tricks the eye into seeing depth where there isn't any. That's right, no 3D headgear required (except the round ones you've already got).

In addition to sounding quite cool, it struck me that this technology is reminiscent of bump mapping in games, which uses a similar grayscale image to trick your eye into thinking there's more detail on an object than there really is.

I think there's some amazing stuff coming in the next few years in computer interaction (multi-touch technology) and entertainment technology. It's hard to imagine the kinds of games we'll be playing in another five years, or on what kind of screen.

A shortcut for modeling people

As a way to settle on my next project, I decided to do some character prototyping. It's something I haven't done in a while, and it seemed like a good time to try a more in-depth 3D modeling project.

The human form is, as you might imagine, a fairly complex thing to model. It's easier with a solid reference drawing, but I'm not a particularly skilled artist either. So, armed with a basic line drawing enough skills to be dangerous, I set out to figure a way to cut some corners.

After some digging for a free alternative to Poser, I stumbled on a relatively new, open-source product called MakeHuman, a free piece of software designed to help people create 3D models of the human form. Turns out it's not quite complete, but I think it's going to be pretty effective for my purposes.

Upon first opening the program, though, my first thought was, Wow, that's quite an androgynous figure. This might not work at all. The model had kind of a masculine face and a vaguely female body. It turns out, though, that MakeHuman takes an interesting approach to body shape. Rather than selecting "male" or "female" when you start, you get to adjust the values along several continua, including male/female, age, muscle mass, body shape, and weight. The default model happens to lie right in the middle of the male/female spectrum.

After tweaking parameters for an hour or so, I had a character that I thought looked decent, so I exported a model to fine-tune in Cheetah. I've been working the last few days to dress it and simplify the mesh.

The only hitch so far has been that MakeHuman outputs a fairly high-resolution model with about 12,000 polygons. I ultimately want one that has four or five thousand, which should provide plenty of detail. Though Cheetah doesn't have a method for polygon reduction, Blender has a good one, so I used that first to get things down to a reasonable level. Despite the extra step, I'm hoping to end up with a decent model that doesn't require building a human form (particularly the head and face) from scratch.

It remains to be seen whether MakeHuman provides a shortcut that's actually shorter, but so far it looks like it will be a huge help in modeling.

Blender 2.46 released

This news is a few days old now, but Blender 2.46 was released on Tuesday. This new version adds an impressive list of features, including:

• Better hair and fur simulation
• An improved particle system
• Cloth simulation
• Render baking for effects like normal maps
• Ambient occlusion
• UV texture editing

I haven't used Blender much in my game development process so far, since Cheetah provides most of the features I need in a friendlier package, but this program continues to amaze me. In my opinion, Blender is one of the true success stories of the open-source movement. It is actively developed and delivers a host of professional-quality features in a cross-platform package, for free. To buy a program with the same capabilities would cost thousands of dollars.

And, by the way, it plays nicely with Unity. Blender is definitely my app of choice for advanced 3D modeling and animation effects.

Take a look at the official Blender gallery if you want to be amazed.