Blog

Amoeba is our first project since we started creating Android games. We have been working on it for nearly 2 years already, and it was just something to start with. Pretty simple and definitely not AAA type. So why the hell it is so long? Here are a couple of random thoughts about problems with developing our first game.

The first thing is skills. I started learning about Android environment without much knowledge about it - application structure, threads and synchronization, memory issues, working with graphics and different aspect ratios, touch interpretations - all those things were unclear. That game had been rewritten three times! From canvas graphics to GLES1.0 because we hit memory overflows due to image assets. From GLES1.0 to GLES2.0 because of features we wanted to implement. And finally global refactoring to merge 5 different Activities into solid one. I am not sure if it was a waste of time, some knowledge was gained during those struggles, but everything could be a bit faster with the right choice at start. Also my desire to work with own code instead of picking some game engine complicates the process of development, but that's personal.

The second is the hunger for features. Both of us are a bit twisted about thinking out some funny cute things and they always chase us. "Let's do that", "Let's add ... here", "Hey, that should be awesome", etc. That featuremania is like a monster on the road of the game development. Especially when it is not your main job and you have no deadlines there. I am holding back with tweaks more often because I have a glimpse about what it would take to implement those, but our artist is much less restrained. Personally now I think that it is more important to have something released than to stuff the game with endless features for ages. But we still continue to do that.

So, after the ..th feature we thought something like "That is already complicated. Let's try something easier". We put aside our Amoebas and started working on Ramble in the Sky. Idea was even more simple, but guess what happened! Features, features and some more features. You can check some examples:
- moving stars at the background of the start screen. And also there are floating cloud menus;
- 2 fall-landing animations with checking if character hits land or water;
- dust animation following cloud landing;
- background zoom and bounce when character is falling.
It is not something big, but they consume time. As for the good point those features are funny and it was interesting to work on them.

As some kind of feature implementation snowball remedy I decided to finish a simple game in two days. We already had gameplay part ready (had some fun playing with match-three tetris clone mechanics) and some assets from Amoeba. There was some extra work with refactoring - that thing was built with old GLES1.0 code. Along with it I had to add start screen, instruction screen, pause and ads. Also I had to improve controls - dropping tetris piece with downsweep while moving with left and right sweeps was a mess because piece column position is too unstable.
All of those took me about 7 (saturday) + 16 (sunday) hours to complete the game Colored Colonies. Despite the goal I still added some unplanned things, but that didn't hurt offhanded deadline much. I tried not to think deep to the core about perfect result as I usually do (thinking, not perfect result :) ), and.. that worryless gamedev process feels nice. We should try something like that with some new gameplay, not just a clone. Well, I like that game, it is classic and cute, but still a clone, one among thousands of those at google play.

And now we are back to our starting game - Amoeba. What is left is some kind of soundtrack and a couple (I hope not many!) of features to finish. So I guess as soon as some music inspiration kicks in we will have that game finally released. If featuremania won't hit us hard again.