UFO: Aftermath Wrap Report

Changes and Enhancements As I mentioned earlier, there were actually very few changes over the course of development and the finished game is remarkably true to the original design document. Perhaps the biggest change was the decision to limit the amount of weapons and to allow the player to manufacture instead of "develop" them. Originally, we thought the player would have an unlimited quantity of any weapon he comes across - once you find a weapon in a mission or develop it in your laboratories, you have an unlimited supply; after all you only need seven of them, so why keep count? Because of feedback from beta testers we decided to change this, and in retrospect, this was a good decision. Some players complained because it is not "realistic", but many more were satisfied with this addition to gameplay.

Major Challenges The single biggest challenge was obviously that year without a publisher. This is not about the game itself, but I cannot skip it. It was a real test of our team's cohesion and determination, and I am proud we passed it.

Other than this, the biggest challenges were the run-time generated random missions. We wanted to have random generated missions in the game because it adds to the replayability, and also because it was a feature of X-COM games. This is not that easy in 3D though, because a 3D scene requires certain things for effective rendering, be it BSP trees or lightmaps, that take a long time to compute. Few players would be willing to sit back and watch the "Preparing mission..." progress bar for more than 30 seconds. The problem was exacerbated by the fact that UFO: Aftermath takes place on Earth, and the player naturally expects a mission in Africa to look different from one in Northern Europe.

We chose a two-pronged approach to this problem. For outdoor missions, we had true random-generated scene (based on the location on the globe, we selected type of surface, relative flatness and hilliness, vegetation, etc.), then calculated a simple lightmap to make the terrain features more pronounced. In cities, bases and UFOs, we took a different path; the mission is assembled from blocks, which themselves are assembled and lightmapped by designers in advance. We have six architectural regions. Each contains about 30 blocks, and each block then contains three lightmaps (for day, night and dusk / dawn).

It turns out in the actual game that the outdoor missions do not work as well as the other ones. They look blander and they are not very challenging tactically. This is despite the fact that an enormous amount of work went into making their generator "smart". We succeeded in creating missions that look believable, but we were not able to make them as interesting.

On the other hand, the city and indoor missions work quite well. The cities have atmosphere, and the bases are probably the most rewarding tactically. The UFO missions are little bit cramped, but they look suitably alien.

Best Decisions I already mentioned what I think were the two best decisions. The first one was to go straight for completion, never writing code that we knew would have to be replaced or creating art that would have to be thrown away. This was quite difficult at times, especially when we were looking for a new publisher and the game we were able to show was but a shadow of its future self. The second one was the decision to limit the number of unlimited weapons, for the reasons described earlier.

Key Strengths Actually, if I can consider this topic together with the next one, I feel much better speaking about areas for improvements than about key strengths. Is this natural? I think it is; after all, we of all people know best where we cut corners, what could have been and is not.

Still, there are quite a few things we are proud of. I already mentioned one, the way we handled development and how we almost hit our deadline. The other strong element of UFO: Aftermath is SAS, our combat system. Many things about the actual implementation could be altered or improved, but the core system works exceedingly well. It is fluid, it gives you complete control, it waits for you when you want, and doesn't ask you to wait for it. It also feels quite natural, the way tactical games are meant to be played.

Another thing we have a good feeling about is the game's stability. Sure, there are crashes and there were bugs, but the crashes are almost all Windows-related (problems with drivers) and bugs were few and we managed to catch them quite quickly. The feeling goes beyond these petty details. We know we have a stable and robust application we can expand and modify. We will be able to re-use quite a bit of its code. This is not the feeling - I hate to admit it - we had when we finished our previous project. Whenever Original War stopped responding on my computer, even for a fraction of a second, my heart missed a beat, and I certainly expected it to crash. Whenever UFO: Aftermath crashes on me (very rarely) I am sincerely surprised.

And finally, I would mention the game's consistency. There are no jagged edges. If we had another year, we could improve almost everything, but the game as it is now is very well balanced - not only gameplay-wise, but also from the point of view of the interplay of its various elements. It is proving very addictive and the feedback from the players is our best reward.