Following the launch of our first title, Bun-Fu: Way of the Rabbit, we have a new project on the horizon, and this time we’re going to see how far we can push the humble Android with a glorious 3D platformer, and this series of posts will follow the development process as the game comes together.

So what did we learn from the development of Bun-Fu? Actually, quite a lot – not because it saw any success, but because we made every mistake in the book. While the game received some genuine praise from users, it also got some 1-star reviews from people who found the controls buggy on certain devices, which pretty much killed the launch from day one.

Another mistake was using the now legacy OpenGLES 1.x, which was intended to shorten the development time that would otherwise be spent writing shaders etc.., but being so limited, this decision had the opposite effect, and meant the game engine couldn’t be easily adapted for future projects. For a simple platform side-scroller, 7 months in development is a little excessive.

So as well as getting the controls right from day one, our next game will be using the much-more capable OpenGLES 2.0 with a brand-new engine written from scratch.

If you’re thinking “Why re-invent the wheel, when you could use an engine like Unity?”, the answer is simply that while third-party engines are very capable, they offer less low-level control of the rendering pipeline and tend to be bloated with one-size-fits-all features; when mobile games in particular benefit from a bare-bones approach. The downside of writing a proprietary engine is that nobody else has spent months perfecting things like collisions, draw-call batching or any of that dreaded matrix math, so it all has to written in-house before you can start the creative stuff.

Luckily, we took care of the basic graphics engine last year, and it’s time to make a start on the game itself, which will be revealed in the next post.