Christopher Dai's Visual Diary

Contributions

Concluding on this computer game modeling project. I will run through brieftly my contributions to this project. The concept of the game is to have 3 different main farm heroes and 3 different animals with each member of the group modeling 1 character and 1 model each. My characters are a chinese farmboy and the evil chicken. The following are my initiate ideas.

Modeling

When the basic concept design has been finalized, we began to allocate our roles. I undertook the role of the primary modeler. As such, i undertook the task of modeling most of the elements in our scene. Besides my main character, the first 2 major assets were the farmhouse and the tractor. I actually went on to design these assest first as I felt it would take much lesser time to do. The initiate models I made were based on the early designs. The focus at that time was to create a really low model, however as it turn out the lowest triangles weren't the prettiest. I went on to add a reasonable amount of triangles to shape the model further in my second attempt. I felt that I could push the design further without adding too much triangles. I was quite satisfied with the final result I ended up with.

I actually assigned a basic colour scheme to these 2 assets. Unfortunately, the colour schemes became the final look as I couldn't finish apply texturing them in time. It was quite a shame.
When I am done with the two assets, I moved on to model my own character. I had a plan in mind to possibly explore some relatively new game models techniques used a lot recently called Normal Mapping. I did some test before I embark on this process to see if it will work. The test result came out fine.

As such, I started building my model on the high side before eventually reducing it down to a game level model. It was quite a time consuming process, which I soon found out because I had much more vertices to handle with during the modeling and UV laying process that I need to do before I extract the normal map.

The result for my model didn't turn out quite like the test. The lo - res model could not display the normal map. I believed it was the UV layout for the lo - res model that caused this problem.

Since it was an experimental process, I didn't want to spend too much time fixing the problem.
After my main character was completed, I went on to model the evil chicken, the game environment and the props in the environment.

The overall models for our entire game were quite efficient as we manage to keep it under the 10,000 triangle budget.

Texturing

As for texturing, I supervised the team in the unwrapping process for all the models in our game. We had quite a number of prop in our game so I thought of creating a workflow that would make that process more flexible. I found that the Reference editor of Maya is quite a great way to work with because that would allow us to work independently while still building towards the finished project. All the props, terrain and characters were first created in their own master file before I called them in as references as I put together our level design.

In this way, the rest of the group could still work on the textures in the master file, while I put together the scene and layout the elements in a reference file.

Problems / Conclusion

As with most projects, time management is extremely important. I think our game had quite a good concept, which allowed us to create many different characters and prop. That was also the bane of our project because that would mean more elements to manage. I think we would have been able to create a much better final product if we had better planning during the pre-production stages. As for myself, although my normal mapping process didn’t work, the process was quite enriching as it allowed me to understand the technique better. Despite all the hiccups, I think we did manage to get a final product that quite match what we wanted to have in our concept.