Production Diary: Prototyping & Research
Updated: Nov 29, 2022
Week One: Briefing
This week, we were briefed on the module and the tasks (which we have been referring to as "Challenge") we will accomplish. We have two practical challenges to complete: one challenges for October, and another for November.
Each month we are to pick a challenge from one of two provided challenge sites ("The 11-Second Club" or "AnimChallenge") and upon choosing one for that month, we are to follow the challenge's guidelines to create a short performance based on the challenge's theme for that month.
Since the Challenges will be announced at the beginning of each month, I will be focusing on the October themes at the moment. The 11-Second Club's theme is an audio from "Ghosted" (TV Series 2017) where two people have banter in discussing a bat man. The AnimChallenge's theme is "Small Animator, Big City" where they use a Black Cat rig, Little Cat rig, and a Pig rig exploring a big city.
I feel drawn to the AnimChallenge "Small Animator, Big City" challenge because I am very drawn to stories which don't require dialogue to explore the story. I feel as though exploring a creature rig may challenge my skills and comfort zone as an animator. I may change my mind early into the following week, but for now I find the "Small Animator, Big City" prompt very appealing.
The rules and guidelines for the challenge:
Maximum 10 seconds.
Any rigs, any medium, any style is okay.
No plagiarism, but inspiration is okay.
The animation must be submitted on 31st October.
Upon seeing the video announcing October's AnimChallenge, I had several reactionary ideas:
[Day] A cat racing across a busy city street to avoid getting hurt.
[Night] A cat carrying its kitten in its mouth, but scared into a dumpster to avoid humans.
[Day] A pig running down a sidewalk, bumping into pedestrians.
[Night] A cat running after its kitten.
A kitten settling in an unconventional resting spot. e.g. an unreachable corner of a baseball stadium.
Overall, I am very interested to see the potential for these two challenges.
Week Two: October Challenge - Storyboards & Animatic
This week, I finalised my competition: the AnimChallenge's "Cat" Challenge. I had a brand new idea where the cat would be in the middle of the street and escaping
I set the objective to complete a storyboard and animatic of the story. Unfortunately due to circumstances outside of my control, I was not able to finish my first storyboard. [Find the incomplete version below.]
This idea was inspired by my reaction. So I searched for reference footage and found some from Eadweard Muybridge's book "Animals in Motion," which captures the extremes and contact points of the cat.
After speaking with my tutor however, I realised I misinterpreted the challenge. The video on the AgoraCommunity website directed the prompt as "Small Animator, Big City" where I understood the prompt as showcasing the size between a small creature (such as a pig or cat) and the big city. The actual prompt was "Become a kitty and so what you do best! Be adorable, cause mayhem!" and the previously mentioned video actually reiterates this through example animations of the cats participating in fun and slice of life activities.
In our discussion, the "Cat Running" story I attempted to tell seemed to be too dark for the challenge or what the Challenge was looking for; in addition to the cat not causing any mayhem or being adorable. This misinterpretation on my end required me to rethink my work and what the challenge was asking of me. Back at the drawing board, I wanted to still incorporate the cars in the scene since I found such interest in the cat interacting with this sense of danger.
I attempted some ideas and landed on an inverse of my original idea: instead of the cat being scared of the cars, it would rather be pompous with a cocky attitude walking on the road. The cat would walk onto the road, and the car would notice it and come to a halt. It would stop so abruptly that other cars behind it would crash into it continuously as the cat smirks and giggles at the cars' misery.
I showed the storyboards and received a bit of criticism of the story emphasised on the cars instead of the cat, and I should rather transfer my focus to the cat instead of the cars. Upon the greenlight, I decided to focus on that in my work next week.
Week Three: October Challenge - Blocking
This week, we focused on blocking. I found a car and street assets from Animator's Starter Pack; and the buildings were a render of the stairs from "The Joker" (2019) from CG Trader.
I set up the scene and imported the cat. I found an issue with the rig as it did not scale properly, and created some terrible contortions. So I made the assets in the scene larger instead, yet the assets were not visible after a certain distance; yet it was solved by increasing the distance of the far clip plane. In my layout video below, you can see the car appearing out of nowhere in the scene causing issues with how the story was being told.
Using the video above, my main priority was focusing on the cat. So the cat follows the cat throughout as the cars crash into one another. I shifted perspective of the cat for a three-quarters/low angle view for the cat's smirk/giggle. This was because I wanted the audience to connect with the cat, and by placing the camera on the cat's level it would help the audience feel more attached to it.
Doing this in layout, I realised a smirk/giggle wouldn't fit right with this; so instead I decided on the cat looking at the camera cunningly (breaking the fourth wall) which fixed the scenario. My major issue I realised was my timing, since I was doing a rough take instead.
Upon review from my peers and tutor, I was informed on the cat being too simple I should add complexity to show my skills as an animator. Such as adding a jump in the beginning, when the cat is jumping down from the sidewalk to the road; or the cat sitting in front of the car, taunting it almost.
I looked at Eadweard Muybridge's books on "Animals in Motion," although it provided helpful images of how a cat moves, it focused on a cat running instead of a cat walking. Searching online, I found relevant clips for my reference footage and compiled them:
When I began my blocking, I realised there would not be an effective way for the cat to sit in front of the car and get up to walk after making the car crash. The pacing and timing would work, but not in a ten second limit. So instead, I thought a more effective way to show the cat's intent is to have it freeze with its paw up, staring at the car; due to the cat's body freezing, it could be interpreted as fear instead of confidence. To counter this, I found footage of a cat's tail steadily and casually wagging to show happiness and curiousness; I kept it side-by-side with the actual walking reference to avoid any timing confusion or technical issues in the future. The final reference footage I used for this challenge is below.
My first step in blocking was to achieve the Cat's walk. There were several steps I accomplished with this:
1) The Cat's Legs' Key Poses and Local Position to the Environment (and Car)
2) The Breakdowns, Foot Contact, and Pose Polishing
3) The Cat's Tail
4) The Cat's Weight (via Counterposing)
5) 12 Principles Polishing (Squash & Stretch, Secondary Action, etc.)
I felt really happy after completing this since it looked really good even without splining and polishing. From the tail to the breakdowns, I was really excited for this. To wrap up the work, I offset the tail controls to make them appear more natural. I did a few tests and really enjoyed them.
However since there was so much blocking and breakdowns, I spent most of my time on this (a week and a half as opposed to the initially planned one week) so I would have to go through splining and polishing very quickly.
Week Four & Five: October Challenge - Splining & Polishing
Since splining and polishing were done around the same time and I continuously polished as I splined, I decided to merge these two sections together.
I attempted to do several rough splinings. The one I was quite happy with was the clamped tangents polishing, as it proved with such an appropriate movement of filling in the gaps between the frames.
I wanted to polish the clamped tangents, however I encountered an issue: my rig's controls didn't connect to the Cat's Geography, so I was unable to add new key frames. So I could not create new key frames or breakdowns, which caused a new issue with my inability to express the cat's emotions through physicality. I tried to change the cat's ears for an ear flick, but the key frames wouldn't save. Neither would the cat's expression change when altering their emotions. I wish my expressions were a lot better, but due to the time limit and my lack of understanding of how to fix this issue despite an extreme amount of research proper methods to fix the issue. I settled with little expressions, but in the future I will solve this issue.
If you see in the video below, when I directly select the controls, no key frames appear or can be made. However, when I access the controls through the QuickSelectSet I made early into the project, the key frames appear. I cannot select individual controls but rather the whole set which caused the issue.
When I first splined, I found the control for the jump was messing up and the poses were glitching. I was quite confused as to how that happened; I realised I put my circle control in Stepped and the poses were remaining at their position. So I made the graph lines linear between each of the points and that seemed to fix it.
I wasn't too happy with how the car rendered, but I believe that was due to my own error. I duplicated the car and didn't separate the two assets. In addition, I realised the car crashing into the car may seem abrupt and a surprise to the viewers. So to clarify the surface the cat is jumping on is a road, I added a second car driving on the other side of the road.
I was really struggling to figure out how to show the cat's intent to crash the car, especially since I was unable to add facial expressions effectively. So I showed the cat sniffing the car, walking forward and looking into the camera: as a fourth wall break where the cat is looking at the audience. It was a good idea, but not with the time presented. If there was more time in the video, it would be more effective.
Upon further research, I found some interesting tips from Richard William's "The Animator's Survival Kit." He notes that quadruped animals have their own specific "up and down" movement when it comes to walk cycles. The head, chest and respective shoulder of the animal goes down as the front leg makes contact with the ground, while the pelvis goes up when a back leg goes up; and the same is true for the reverse. Unfortunately, I found this information after the submission.
In addition, another criticism I received was the offsetting. I realised apparently I didn't offset the key frames too well and therefore it didn't appear to be there in the final product. After consulting some of my peers, I realised my offsetting was on a decimal frame and not an integer frame; explaining why it was not noticable.
Week Six: October Challenge (Review) & November Challenge (Storyboard/Animatic)
For the review, we went through the animation. All bases were covered in my work, my time management and goal setting was effective; and even if there were moments that were ineffective, I held myself accountable. My tutor specifically spoke about my delays: specifically with redoing the October challenge, and delaying the work to develop the idea. They also brought to my attention that even though my piece was good, it would have improved with more filmic language. I took note to do add that for November.
Chosen Competition for November: 11 Second Club Audio
Text: "Even if I'm not real, this moment is. Right here right now. This moment is real. I mean what's more real than a person trying to help someone they love?"
The above mentioned audio is from "Free Guy." I have never seen the film, so I am safe from any viewer bias. I will not be looking for the source of the audio to avoid any scene clashing.
I had several reactionary imagery to the audio:
Main character falling from the sky
A robot talking to their creator
Slow Pull Out from Close Up Shot to Extreme Wide Shot to show their facial emotions to the destruction of the world around them.
A person falling off a building after letting go of their friend's hand; sacrificing themselves to save the person they love.
Two people looking at a sunset. Person A is sad about Person B, and Person B (the audio voice) nudges them and opens their eyes to the sunset.
Person A holding a shield against a blast, and protecting Person B.
I decided to break down the audio and what made me think of the story like this:
Character is "not real". Like a robot or some kind of digital machine (a la Pantheon)
Despite being "not real," character still has sentience and emotions; experiencing life.
They love a person (further confirmation of sentience) and they are helping that person.
Delivery is quite dramatic and somber, almost as if they're making a sacrifice or reaching a dead end climax to their arc.
They are not thinking about the future, or the past, and are basking in the moment. (Typically, this means an event is occuring in the near future which may disrupt their "life.")
They do not care about the permanence of their existence. Rather, they prove themselves to be real based on their emotions and how their ability to help those they love is the most real of all.
After some experimenting, I found myself drawn to (5). Below is reference footage of my explorations of different movements to (5):
Eventually, I chose one clip which I felt would be most effective in what I wanted to portray:
Based on this, I created a rough animatic to showcase my camera angles and idea.
Upon review, I realised I was rushing with my idea. I was thinking about movement of the characters but not the actual story; and when I started to think of the location of the characters, I realised the sunset idea wouldn't be effectively communicated through the scene at my current level of expertise.
I decided to quickly redo the idea on a new level. I was brainstorming ideas and looking back to my initial reactions. I realised what I was drawn to from the audio was the discussion of impermanence of the self; the way the character could disappear at any moment, yet finding the joy in their existence.
I recently was studying a show on AMC+ called "Pantheon," and I really connected to the sci-tech horror setting they were presenting and I believe that shows in the concept. I was drawn to their Uploaded Intelligence and how they brought up the philosophical concept of the self with their impermanence in their technological state of being.
To showcase that, I wanted to convey the idea with TV static: as often television remains in a state of constant alignment, yet television static reminds of the impermanence of the state of television. In order to show this conflict, I wanted to show two sides to this argument: the person who has accepted the impermanence of the self and is lost to it, against the person who cares for them deeply and doesn't want to let them go.
I made an animatic to conceptualise this.
Upon showing that to my tutor, it was reviewed once again. A few notes to make: the character (we will call him "Eleven") was placed in the scene and the audience could feel his 3D presentation as a character through a window instead of a Television. I was informed the first thing to do is to import his animation onto an image plane of sorts in order to convince us of that.
Another thing we had to note was the vagueness around the idea. I tried to showcase this in some reference footage, but in order to clarify the movements, I recorded another animatic using reference footage to showcase this.
And finally, the biggest thing to do was have several visual indicators of what is going on between the characters: since the story is 11 seconds and requires for someone to understand the full concept within that time.
Week Seven: November Challenge - Lipsyncing
I got straight to work on my lipsync animation, but then came the facial animation. I wanted to go for a more soft feeling of acceptance for Eleven, almost as if he were whispering the words. I wanted Neka to have a harder expression; she wouldn't be speaking so she would need to convey her emotions through facial features alone. I recorded reference footage to show this.
I have nothing much to say about the facial reference. I was enjoying this since I recently understood how to do facial animation references. I had a bit of difficulty with the tongue, however I think I resolved that issue.
Week Eight: November Challenge - Stepped & Splining (Eleven)
This week, I focused on Eleven's Animation. I spent a bit of time tweaking the facial animation here and there.
I wasn't too sure how to handle his body animation, so I attempted my reference. However, I realised I was trying to do his upper body and didn't record the reference for it, but just the head. I attempted several passes of reference footage, but I wasn't quite happy. So I settled for one that I liked the best of all.
I'm not sure what exactly I didn't like, but I still wasn't too happy with it. I knew having a completed good animation was better than consistently polishing an unfinished work.
After completing the animation, I applied a green screen background, transferred it to After Effects (where I removed it and applied the static background) and I was ready to import it into my new file for Neka's Animation. This was actually pretty fun work to do since I've never used a green screen and such effects in After Effects so learning about it was actually really intriguing. One thing I would change if I had to redo this was to remove the strange green tint during Eleven's animation; clearly a result of the green screen, however with my limited knowledge I was fine with the work.
I downloaded an asset from CGTrader for a TV and a Room, and I was able to apply the sequence to the surface to make it actually convincing that it was a TV. I was super excited since I has never done something like that before. It was super intriguing to learn how to put it on a specific face of an object. I imported Neka into the scene to set up the staging.
Week Nine: November Challenge - Stepped & Splining (Neka)
With Neka's Animation, I realised my big issue I was experiencing. I have only recently done animation with characters facial expressions and body animations. Having only recently learnt facial animation, it was becoming difficult to balance the two. In addition, it was my first time working with two characters to animate in a scene, so I had to pay a lot more attention to my time in doing two animations simultaneously than I previously did with the October Challenge where planning took longer than anticipated.
I experienced the same issue with Neka that I did with Eleven: a difficult time choosing a reference footage I liked. The reference had to be short enough to keep the focus on Eleven due to his non-permanence, but long enough to express Neka's emotions at losing Eleven. Yet again, I settled for a reference footage and, much like with Eleven's animation, I wasn't liking it all that much.
Neka's animation mostly relied on one specific posing, a bit of hand animation, and mostly facial expressions. It took some time, but far more than I expected. I did a first pass at Neka's facial animation with the completed draft, however the movements were too quick and did not pause on one or the other.
I completed a second draft and liked that better in time for the deadline.
In terms of the camera, the environment I chose was a room too small for what I wanted to convey. In order to show distance and not have the character too close to the screen I adjusted the aperture of the camera. This made Neka appear further than she actually was, and also made the room seem larger.
I distorted the scaling of Neka's playblast in After Effects to give the idea of contortion, and technological distortion. The initial playblast (RoughFinal) was an example of keeping to regular scale, while with the final video it showed the effect I wanted better.
Comments