FMP – Final Video
Finally, I managed to composite the full video for my FMP. Since it doesn’t require any sound effects, I didn’t add audio or background music. However, if possible, I may include music in the future. As I mentioned in previous posts, the title of this project is “Invisible City” (It was Invisible cities in concept stage, I just changed it to Invisible City)
Outcome
Breakdown
Critical Reflection
Compared to last term’s project, this one felt much more spontaneous and unclear. Last term, I had a clear concept and blueprint for how to approach and execute the entire project. However, this time, I wasn’t certain about the direction of my project until the very end. The concept—derived from my personal reflection on the overuse of social media in daily life—is something I believe is strong. I faced a key decision early on: whether to focus on creating fascinating visuals (FX) or exploring a variety of scenes (compositing).
Since this was my final project for the course, I wanted to include sophisticated FX and visually impactful elements. To achieve this, I should have spent more time refining the concept in the early weeks. However, due to various reasons, I struggled to concentrate at the beginning, leaving me feeling confused and unsettled throughout the project. When I finally decided to prioritize compositing with some FX, I designed scenes to showcase both. Despite my efforts, I feel the final quality doesn’t meet my personal standards.
The filming process was another regretful aspect. Without proper planning, I filmed the footage without fully considering key factors such as reflections and color consistency. Additionally, I filmed the green screen footage without using tracking markers on the screen, which caused significant delays in my workflow. For example, when I reshot scene 2, the colors were vastly different from the original footage, leading to time-consuming color grading. Even after all the adjustments, I feel the results don’t integrate seamlessly into the scene. There are countless aspects of each shot that I wish I could improve, but the limited time frame and constant pressure made it difficult to address everything.
The most regretful point, however, is the color grading. Knowing that color grading is not my strength, I should have paid more attention to it. Unfortunately, the delays in the rendering process pushed back my timeline, leaving me with little time to focus on it. This was a mixed mistake on my part, as I should have prioritized finishing earlier stages to dedicate sufficient time to color grading. As a result, the final video suffers from inconsistent and subpar color grading. If I had been able to add a touch of haze to match the title, Invisible City, and create a dull, misty look while maintaining a consistent tone throughout, the overall quality of the video would have been much stronger.
This project also marked my first time using Cinema 4D, and as a result, I faced many challenges due to my lack of familiarity with the software. Rendering was especially problematic. Since LCC’s render farm doesn’t support Redshift or Cinema 4D, I couldn’t use it for rendering. This limited my ability to run enough test renders. Although each frame took only 2–3 minutes to render, the final scene required nearly 600 frames, which took almost an entire day to complete. For the final scene’s disappearing effect, I encountered an issue where motion blur wouldn’t render correctly, even though the setting was enabled. I still haven’t figured out the cause. Due to time constraints, I had to leave the scene without motion blur, which made it feel unnatural.
Through these challenges, I learned a great deal about identifying problems and finding solutions. Sometimes, I relied on forums or AI tools, and other times, I had to troubleshoot the issues myself. However, there were still cases where I couldn’t find a solution—such as in scene 4 with the emoji particle shot. Despite baking the simulation, errors persisted, and I had to render it frame by frame. Although these errors were frustrating, they provided valuable hands-on experience with Cinema 4D. I now have a much deeper understanding of its features and potential for future projects. While Houdini is incredibly powerful, its coding-heavy workflow remains challenging for me. In contrast, Cinema 4D’s interface felt more intuitive, and I learned a great deal through the simulation tests I conducted during this project.
In the end, while the video is complete, I recognize there are areas that could have been improved. However, the knowledge and skills I gained throughout this process are invaluable. I believe these lessons, particularly in troubleshooting and problem-solving, will be extremely helpful in future projects. Although I can’t say this project was completed perfectly, it was an excellent learning experience, allowing me to identify my shortcomings and improve for the future.