Spring 2013 in Review: Part 1 - Raaargh!!!...

Another notch in my educational belt and a step closer to being done with school. Just 7 months to go and I'll be embarking on a completely new phase of my life. But where am I at with the 5 different projects  I started at the beginning of the semester?

RAAARGH!!!
My primary educational and professional goal this semester was to produce a fully 3D mobile game of giant atomic monsters. Here are some of the pics from the game demo we put together:

KARKARIAS

FTAGHN

THE COMPLETED USER INTERFACE (NOT YET FULLY IMPLEMENTED)

GAS STATION

WHERE THE ELITE MEET TO GET 2 FOR 1 PIGGIES

EDGE OF TOWN

SCENIC DOWNTOWN

KARKARIAS ATTACKS THE TORSON SPACE CENTER

And here is a game play trailer we slapped together...

RAAARGH!!! TRAILER 

Now, as far as we got in the Spring Semester, there is a lot of work still to do, mainly because the production of a mobile game using UDK fo iOS is a major pain in the butt. Seriously. Here is a small list of some of the bigger issues we ran into while working with this technology:

1. Bone Limits
Mobile phones don't like the bones! And for those of you who aren't in the know, bones are exactly what they sound like for a 3D model: the skeletal structure that makes it possible to animate the model. In the case of an iPad 4, you are pretty much limited to a total of 70 bones, which isn't a lot for a single model much less two or three. We ened up art around 30 for Karkarias (the shark monster above), but adding in Ftaghn (the squid monster) was going to push the upper limit and that meant there was no way to have battles with 3+ models in the current build. We considered vertex animation for a future build, especially for things like Ftaghns mouth tentacles, but didn't have time to really implement that properly.

2. Actor Limits
Actors are individual models in this case and our format was limited to about 20 of them on camera at any one time without bringing everything to a screeching halt. Now, when you consider every building in the city is destructible and has to be a separate model, that's a lot of actors and requires a lot of fancy camera work to keep the actor count contained to a low number. We came up with some workarounds for the demo, we built all the rubble from the destroyed buildings as a single actor underneath the existing buildings for instance, but these will not work for the final product.

3. Lighting
Shadows don't work for iOS even on the iPad 4. They flicker in and out and generally look goofy. Dynamic lighting and particle effects are also huge drains on the tiny tablet processor. That is why Karkarias doesn't use his laser beam eyes in the current build.

There are a horde of other problems but the big picture is basically this: the technology is not there yet and UDK is not the way to do this game even if it was. This game, with its indirect control scheme, destructible terrain and monster building requirements really push the limit of what it is designed for: to make good first and third person shooters.

So in the end, the project was a success in that we really pushed the current tech and learned the limitations of the mobile platform. It was not a success in the 'getting a game to market' sense, however and while I intend to finish it, I'm going to wait until the summer to see how the technology evolves. And then we'll rebuild the whole thing from the ground up in another engine, or possibly even create our own.

PROJECT STATUS: ON THE BACKBURNER UNTIL FALL 2013.