Saturday, 2 March 2013

Student Project IV - Post Mortem

The fourth student project at the Games Academy is at an end.

It was quite a challenging project, unfortunately not because of high technical difficulty but because of endurance and staying focussed. We struggled to find a vision we all could live with and could work on, which resulted in a lot of unnecessary work and motivational problem for most team members. We found our vision about three quarters in the given time and had all to work overtime to finish it. I initially wanted to work on Cookie X on the weekends but there was not time for it and I even had to cancel on parties and seeing friends.

The problem was that a helicopter is not really suitable as a vehicle in a racing game, so the game designer played around with different mechanics until he found the "taxi mode", where the player has to transport items from one place in the room to another. Ideas like a rescue helicopter or a real taxi helicopter were tested but not fun enough for us. We wanted something different. And that we found.

Why not build a city out of building blocks via helicopter. That idea was kind of strange at first but combined with a mechanic to pick up blocks and a kind of 3d blueprint of the building to build, it worked. The player controls the helicopter, flies around the room, can pick up blocks of different shape and size that are distributed all over the room, fly with them to a designated building place and drop the block at the right spot in the blueprint.

To accomplish that we build the mechanics and made play tests but the pick up process and the placing inside the blue print was quite difficult. With a target group of casual mobile gamers in mind, the controls were far too difficult. After a long period of time (one programmer being absent helping his wife with their new born kid), we managed to fix even that and in the end we had a nice little casual on the android platform.

Like I mentioned in the previous post, I worked on the UI, menu and asset integration. Unity has some problems with 3rd party revision tools like SVN, so we had some difficulties there but after some mistakes we managed to overcome that obstacle as well. Another curiosity is that 3d models imported into Unity are scaled down to 0.01 of their original size by default. There is a scale option but while importing a bunch of models it was overlooked, which had unwanted results and I was the only one charged with integrating all assets after that. For a mobile project we needed to limit the draw calls as much as possible but if there are two instances of the same model (which use the same material) inside a scene but they are scaled in Unity they will produce two separate draw calls. So, all models had to be scaled outside of Unity and the material had to be versatile enough to be used on as many models as possible. Unity is able to combine meshes that use the same material and that are switched to static (if it is not moved) reducing the draw calls and thereby boosting the performance on mobile devices.

Due to the short time frame at the end of the project we could not implement all the needed tweaks to make the game work with older or less powerful devices than my tablet but we will do so during the next few weeks.