Four weeks have passed since I placed the first polygon of this project and now everything have been implemented in engine and is ready for play.
It has been an interesting challenge to do something of this proportion in such a limited time and I have forced my self to learn so many new things, both artistically and technically.
The challenge for me was to build a level of as few modular pieces as possible and while doing so getting and understanding of the advantages and disadvantages of modularity.
To build this level I had to model four main pieces, some of these where then altered to add variation.
So in total seven models and four textures.
Except from the modelling and texturing I also created my own shaders and built all of the light as well as everything else that is required to make the level playable with bots or over lan.
One thing that I'm not pleased with is the reflections in the metal. I never managed to pull of a decent cubemapshader. My result is in the screens below. This is something that is on my to do list and will be fixed as soon as I get some time on my hands.
The level itself is not the most fascinating Capture the flag map ever made but this was also my intention.
By making a small non-experimental level I managed to mitigate my risks and instead give the project a bigger chance of being successful, which for me was more important.
I would much rather work on a smaller scale, making almost certain that I would manage to complete all of my planned tasks, than making something half-hearted that I would not be able to stand for.
After this project I have learned that you should use modularity for almost everything, the question is rather how modular it can be. If I where to do the same project again I would also concentrate at making the textures modular also and rather build my set pieces from the textures instead. By doing it that way you can save precious computing and instead spend it on where it matter e.g. characters and first person weapons. This will probably be my next project when I find time for it.
Another drawback of working with modular pieces in lightmap based engines such as UDK is that it can produce seams while tiling meshes. These seams are produced if the UV seams is placed in the middle of a pixel. A smart way around this is to model seams where these seams are produced, you can see this in my floor piece.
The result for the curious:
|The four main meshes|
|Print-screened directly from the editor|
|You can shoot, also|