Senior Production: City Concept Test

Early development

Senior year is often the time for multi-sized group projects in order for the graduating class to demonstrate what they have learned during their years in a program. In my case, truly wanted to test myself and push the boundaries for what I could do. This lead to the development of a solo project in which I would develop a full city scene simply based on one piece of concept art produced by Zhichao Cai.

The initial concept showed great potential for interesting forms with subtle hints of "wonk". This was the primary reasoning being choosing this piece as working on a project for a full year requires both detail and scale. With this piece providing that and more I began work on the early blackouts and overall layout of the scene.

While initial Blockout from the reference was fairly straightforward, the rest of the scene immediately gave me issues as I had no clue what the rest of this city would look like. I only had a rough visual design that came with the concept art stating that this image comes from an overgrown city somewhere in fantasy China.

Moving forward I quickly realized that this project was not only complex from a technical side, considering I was one person working on the whole thing, but it was also complex on the creative side. I had absolutely no clue what I really wanted this city to look like once I was done and early blackouts quickly were replaced over and over.

After several iterations, I ended up with the idea of a railway going through the city. Coming from Chicago I really enjoyed the way the rail system lived with the city and interacted with everything around it. I wanted to create a city that was once full of life with all of its flaws and inconsistencies. Overgrown and left abandoned, this would be a dead city but a thriving natural environment. I knew early on that I would not have time to create the small animals or even the steampunk vehicle that we see, though I would love to tackle it one day.

My standard workflow at this point was to take each building one at a time and add "wonk" and details by hand. Unfortunately with the level of detail needed for each building this quickly became a slow and time-consuming way of production. While the results were somewhat satisfactory, I needed to quickly speed up my workflow.

Module-based workflow

At this point, I was reaching out to other classmates, graduates as well as my professors in order to find out what my options were. After fishing around for ideas a former graduate of the program introduced me to a modular workflow. While I had never heard of working this way before, I desperately needed a faster workflow so that production could continue at a steady pace while also tackling other classes and life. We always joke about the "Buss factor" while working on projects but this project itself is the buss.

Early iterations of working in modules were, in all honesty, a failure as I did not understand what made a good module for intricate buildings. Basic shapes provided nothing at all for the complexity of the scene that I needed. While they served their purpose of filling the scene, it was not enough.

Thinking ahead on how much texturing, lighting and set dressing would be needed for this project, I began to make slightly more complex modular building segments in order to get further variation. The results were successful when full buildings were assembled.

I began to also consider how working in modules could help me set dress the scene as well. With this in mind, I wanted to tackle the rail system that would be going through my scene. Taking a good look back at my time in Chicago gave me a lot of drive and inspiration to create these tracks and how to best use them in the scene with my own creative take.

Moving Forward

A large portion of my time was then spent on creating assets, buildings and everything else needed using this modular workflow. It really shocked me how effective it is from a modeling standpoint. Assembly can be troublesome at times depending on the complexity and overall design of the building but overall it's amazing what you can do with simple forms.

I took some time and went through a few more iterations of the city layout. Overall I was fairly confident in the angles and scale of the project and was ready to move on.

Taking some time after this layout was complete, I had to take a step back and acknowledge that the scale of this project was a bit much for one person and chose to cut back several ideas and removed a few cameras. Just working with some rough lights gave me an idea of the time of day that I wanted for this scene. Taking some time to flesh out the scene had great effects and allowed me to modify large portions of the modular buildings.

At this point, I needed textures in the scene and foliage to cover the whole project. The foliage has given me many issues and at this point, I didn't have any practical results. Textures, on the other hand, were a whole other story. After reaching out to my classmates and brainstorming we worked out a fairly robust pipeline that would help me in the future.

Unfortunately, the world would soon be knees deep in a global pandemic with the development of COVID-19. Our university and program quickly realized how bad this was and wanted to keep students at home. Unfortunately for me, my hardware at home was fairly limited, and as such production quickly came to a crawl. It was heartbreaking to see what I had created so far almost come to an end.

Sitting down with the faculty we discussed how I would be moving forward. Realizing that the scale of this project would no longer be possible for one person working from home with everything going on, we agreed that the best thing to do would be to focus on one shot and attempt to complete it as fully as I could. This required me to step back and really consider my options on how I was going to complete this. In the end, it was about getting as much done as I could with the time I had and with the equipment I could access. While it was upsetting, it was the only option I had at the time.

The remainder of my time was attempting to flesh out areas of the scene with smaller assets, foliage tests for possible future development, and small scale texturing to fill the time.