Description of work done
The assignment required me to use Unreal Tournament 3 (UnrealEd 3.0) to build half of a level, leaving the other part for the person that I was paired with. I went for oriental, Japanese themed layout and environment. Finally, I ended up creating all of the exterior, and the other person did the interiors of a building. The level was finalised, it looked pretty good. The level is fully playable (deathmatch-styled) and only has some minor "bugs". I implemented 3 custom textures to the level as well.
Reflection and Issues Identified (SWOT)
Strenghts: I really filled out the Japanese theme, the level is fully playable, weapons, powerups and health packs are pretty strategically placed. The terrain and environment overall are not boring, lots of changes to the simple "plane" type of terrain, a little forest-like pass, so that the level would look interesting while walking around, and fun to fight in.
Weaknesses: I procrastinated and did my work late, because of bad planning. There are no shadows to be seen in the level after cooking it. I intended putting vehicles inside the level as well, but something was wrong with the terrain. Also, I tried my best to make the lighting look good, but its not nearly as perfect as I hoped it would be.
Opportunities: The level can be made bigger, the terrain fixed. It would really make more sense if the whole building was to be removed, and instead, the rest of the map extended, copied and flipped. It would guarantee equality for both sides, strategy-wise. If the vehicles would work on the level, it could be more interesting and challenging to play. Also, better custom textures could be implemented.
Threats: I could ruin the lighting, and a bigger level means more load on the computer, so I might have to remove numerous static meshes from the level, which would destroy the whole environment that I was trying to achieve. I would say I could corrupt the .ut3 file if i save over a bad action, but since I'm saving the level pretty often and in different files, that is not very likely to happen.
Plan of Action (SMART)
My plan of action would be improving the balance and overall design of the level, which means: 1 - make half of the level and mirror it to guarantee balance for both sides. 2 - extend the map, recreate the terrain so that vehicles can be used in-game. 3 - Implement custom static meshes, to make the environment look as good as possible. 4 - hammer out the lighting. 5 - improve placement of weapons, powerups and health packs.
This is really achievable, and comparing new results with the old ones would take 1 - letting people play for extended durations of time and check if the balance is good, and 2 - try running the first, then the other level and walking around inside them to see if it decreases framerate a lot or not, and of course - looking which level looks better.
The plan is relevant, because it adresses all of the issues identified in the SWOT reflection. All of the problems would be tackled and fixed if I was to create a second version of the level.
This plan could be completed within two weeks at most, if supported by good time planning and no other issues would occur.
Completion/Review/Conclusion
The conclusion is, for the first attempt, I really like what I built. It cost me a lot of time, and most of that time was used not to create the level, but make it work - lots of problems occured with the terrain, cooking of the level, and mostly, shadows. The plan is realistic, and I am sure that acting by it would seriously improve the quality of the finalised product. Basically, it helped me lay down the problems that I have, including a way to solve them, and look at the project from perspective, which makes it easier to understand what's wrong and what's right in the current level.