Cube Orbit – Almost Finished

Before Christmas, I started yet another game project. I had the core game loop prototyped almost a year ago. It always felt fairly accessible as well as fun and from a developers/designers perspective it promised to be a project with limited scope. Just what I needed to finally finish a game, since my last game “Gravity Racing Madness” was released almost 4 years ago. Now it is finally there. The first people play the alpha, I feature freezed; we are still tweaking the right balance, but that is it. There even is a website.

Cube Orbit is a free-to-play casual match-3 single player game with 3D graphics. An average run lasts between 45s and 10 minutes. The perfect distraction for commutes, standing in line, or other short annoyances. It was specifically designed for mobile touch devices. The game has a fairly abstract design with no references to violence, sexuality, drugs, or inappropriate language. It might be problematic for players with photosensitivity.

Digging into colored cube structures is aesthetically pleasing and naturally relaxing. Juxtaposition with a timer leads to a rewarding and challanging experience. The coloring cooldown is neccessary to avoid winning the game solely by fantic tapping and adds another dimension of depth to the game. Carefully choosen bonuses, powerups, and skills allow for long-term fun with the game.

Environment the Third: Considering the Camera Angle and Focal Distance First


The current game environment (see screenshot) has flaws:

  1. Its going for a realistic look which requires lots of texture work and large textures.
  2. Elements like bridges and ladders require lots of tris without producing enough buzz.
  3. The pillars block the view even for ridiculously placed cameras. Which requires expensive z-buffer magic (blending and double draw calls) with complex programming.
  4. The confined environment and camera properties call for a backdrop graphic or other kind of background.

Before I dive headless into the next iteration of the environment, I should do some thinking.

Camera angle and focal length determine how much of the environment can be seen. The angle is determined by the game mechanics: one has to be able to see what one has to see to play the game. The focal length is limited: getting the camera to close will distort the view on the important game elements. Besides these contraints, the camera properties can be tailored for the background.

As the following graphics show, longer focal length and steeper angles lead to less visible environment. This effects the environment elements that can be seen, their placement, and their size. Hence it is very useful to fix angle and focal length before hand, work with a placeholder environemnt, and do all the detailed work on the environment later.


Anyhow, since a foto realistic style is beyond my abilities, would require too much effort, and doesn’t really work for an indie game anyways, I decided to go for a simple, more abstract look. Currently I am thinking about “Low Poly Terrains”. Artists like JR Smith, Alexandre Duforest, Vitaliy Prusakov, or Jeremy Kool are examples for this. It still need to test this might be very suited for my needs:

  1. Based on a first look on a few “Low Poly” tutorials, it seems to be doable with my skills.
  2. Since it is just colors, shadows, ambient occlusion (AO), and some paper/grunge texture. It should be possible to combine many les-res, pre backed AO and shadow textures with a single high-res paper/grunge texture that (re)used for all objects. Should deliver good performance. Needs to be tested to make sure it looks good.