The deadline is here and I have only managed to spend less than 30 hours on the game.

In the end I just entered the sandbox, there was no time to add my menu.

At least the time I did spend working on the game was fun and I did learn a few things:

First @ryan20fun asked what tech I was using and the answer to that is Unity. The whole idea behind this for me was to start from scratch, using no tools or extras, just Unity.

There was a problem with the batch limit, this is as large as a building can be. Adding one more floor causes the mesh merging from Unity to fail.

Shadows is the largest performance killer and I found no way around it that I could achieve in a reasonable time frame.

Second thing I learned is that once you understand how Unity planned on using there Quaternions it's easy to pass math between them and Vectors.

The most of how smooth this game feels is thanks to lerping the vectors and quaternions. The tank as is only took 4 hours to implement and one of those hours was me wrongly using the Quaternions.

The Camera was very fast and because it didn't have any extra behaviors added to it, I got the Battle City feel easy without having to counter code that exists.

I wanted to measure how switching from primitives to complex art is and never found the time for it. However switching from a cube to the building place holders was fast. There was no performance loss at all even if in theory it was 16* the amount of vertices.

I do believe I will not even stand a chance this year, not with the other brilliant entries I had seen. However I did enjoy it and seeing the skills of others as they make browser games, multiplayer games and implement there own engines was amazing.

Postmortem: Game died of neglect.

Edit: Fixed the zip.

ProjectCastle.zip

The zip is 18mb and the game should work on most computers even old ones.