Merritt Lum
Multimedia Digital Artist
Artstation (3D works):
https://www.artstation.com/merrittlum121
Contact:
mgl13@sfu.ca
Date:
February 2023
Software Used:
Blender, Audacity, DaVinci Resolve
Overview
Made as a project of SFU's Sound Design course, requiring to make an original film and sounds to go along with it, ABJECT is an independently made short horror piece animated in Blender and with self-recorded and edited sounds.
​
3D
Given the relatively small scope of the project, I was able to use this as an opportunity to more thoroughly explore the parts of a more complete 3D workflow that I had typically avoided or skipped over in the past for various reasons; most of my 3D projects beforehand, especially character sculpts and the like, rarely moved past initial modeling and basic material assignment, meaning I had little experience with retopology ("tracing" new geometry over a high-density model to make it production-viable), model rigging, or texturing - this project, by contrast, had me engage with all of those steps and more, even lightly stepping into inverse kinematics and animation. In general, having an idea I felt was simple but strong, I was able to work on creation far more than concept, allowing the project to move at a fairly quick pace, only slowing down when dealing with elements I was unfamiliar with.
Project Type:
Genre Piece, Sound Design, 3D Animation
Original mesh for the monster (bottom) vs retopologized version used in the video (top); reduced face count from 257.3k to 4.4k.
Original (left) vs retopologized & normal mapped (+ not very anatomically accurate, admittedly) arm (right) - face count reduced from 91k to 1.4k.
Of such roadblocks, the primary source was undoubtedly attempting to better learn physics simulations within Blender; as my 3D experience has been almost entirely self-taught, I had to consult a variety of tutorials to get the lantern and chain to move as I wanted them to, and often encountered situations where my current method would not support something required, and thus had to learn a new method and adapt what I had to it - for example, I had been using a soft body method for the chain until I tried to animate in a new control point for it being grabbed - something not supported by soft bodies - at which point I had to spend a fair bit of time looking around until learning to use & convert to a cloth simulation method instead, to say nothing of the implementation time.
​
Audio
Though I was still somewhat new to it, sound creation was a relatively more straightforward process; though I was only able to think of foley ideas somewhat spontaneously, it went much more smoothly in terms of actual implementation, and was able to pick up some general notes about recording sound to synchronize with film along the way.
Given that all of the audio sources in the video were fairly natural, I was able to create all of the sounds I needed by simply recording objects from around the house, and sound editing was generally minimal beyond adding basic pitch and reverb adjustments. To list a few, I found that squishing a block of tofu made fairly good fleshy sounds, breaking peanuts in a mortar & pestle worked for creating crunching and breaking sounds, and I ultimately discovered that a keychain with many parts to it, with two sides rotated in opposite directions, was a plenty good way to imitate a much larger chain lowering something down.
The general setup for recording sounds I had; I would record sounds in real-time while playing back the video, and then did any additional sync needed.
Reflections
Given the short timeframe I had, there were some elements I wish could have been refined better - most obviously the lantern's shaking after being grabbed, but also some harder to notice mistakes, and some sounds that are not synced as well as they could have been. Furthermore, I feel I generally spent too much time on the animation, rather than the sound, which was supposed to be the focus of the project; though I feel choosing to do an animated project allowed me to develop my core skillset further, it likely took my focus too much away from the most important aspect.
​
That said, I am still overall satisfied with what I was able to create, and feel the project overall allowed me to both develop my skills further and recognize some areas in which my general project approach can be flawed. On that note, the experience of working on this project also served as something of a reinforcement of the importance for me to learn, relearn, and discard techniques quickly, and to generally be adaptable when working with creative processes requiring technical knowledge; having to struggle between various physics simulation techniques as my needs changed would have turned out decidedly worse if I had stubbornly tried to make my initial approaches work.
Before & After (with some more lighting for clarity).