Month: February 2023

This has been my favourite exercise by far. As soon as I had settled on my video I had a very complete image of what I wanted the animation to look like.

I started by plotting out my thumbnails and when action would happen against the timing of the audio. Based on how I worked on the other lip-sync I drew out and experimented with different mouth options, making a particular choice to only create shapes that were needed and not over complicate the lip-sync. I tried to keep aware of similar sounding/shaped syllables that could reuse existing mouth shapes and how I could make small adjustments that kept the lip-sync fluid and wouldn’t overcrowd it. I also trialed some different design options before settling on a classic cartoon ‘bean’ mouth.

The next step was to roughly timeout the thumbnails and create a lip sync test (below). I found this a much quicker process compared to my first lip-sync exercise and I think using restraint in terms of the mouth shape was a big contributor to this.

After this I worked on fleshing out the movement and adding the mouth shape to the design. I found this far easier that the first lip-sync, partially down to having it done it once already but also due to this being a simpler and more stylised design with less complex movement in the head and face.

My first pass of the full animation came out relatively well. The main feedback was that there was too much stillness in the body in the beginning and end compared to when she pulls down on her face.

Christina suggested initially, to readjust the timing of the audio in comparison to the movement so that the final line “I love halloween” would happen earlier, allowing for a more dynamic performance. I then reworked the body and facial expressions during “look it, look it” to make more extreme poses, added more performance into her hair and a final image that punctuated the ending.

This time I absolutely and resolutely started with thumbnails. I refused to allow myself to do anything else first. After receiving the exercise and working with Ying to create a reference video, I sat down and really plotted out the entire animation.

Starting with the body acting, I plotted out the key frames and facial expressions that would work separately from the lip-sync. I think, working directly with Ying, wrote out the script phonetically and planned out the mouth shapes for each sound.

I started out by creating the body movement with some simple break downs and in-betweens to create a rough version of the sequence. I used the sound file to create the timing of the movement, working with the thumbnails as a reference.

From this I then worked on the Lip Sync separately, creating a mouth shape separate from any body or movement. I used the original mouth shape design sheet I had worked on with Ying and used directly mouth shapes from the reference video.

I then used this as a reference guide to create my first pass at a full lip sync animation

This, absolutely did not work.

It was a bit of a hard lesson but it was clearly very early that this was not a good animation. The concept was there but the lip sync was too complicated and the movement was stuttering and unnatural. I decided, based on my previous experiences with the Body Acting to start the lip-sync again. I initially kept the body movement and first just worked on reworking the mouth shapes and position to create a simpler and more effective lip sync. Looking at the two side by side, it is clear to see the efficacy of the updated version in comparison to the first.

I initially kept the movement from the original animation, but reworked the timing and added further inbetweens and breakdowns in order to create more natural and readable movement.

However this ‘fix’ then highlighted the problems with the mouth itself, making the rest of the movement work only made the mouth positioning and shape seem incorrect. The inconsistency of the placement on the head was highlighted and the face appeared to ‘slip around’ during the movement. Working from helpful feedback I decided to redo the head of the animation, creating a guide drawing to help with head shape and mouth position. This small addition make a world of difference and made the piece far more successful.

The facial acting is probably the exercise I found the easiest to do. I had a clear image in my mind from the beginning of what I wanted the animation to look like and enjoyed the process of playing with exaggeration and performance to create a fun and dynamic animation.

Initially creating the reference video was quite hard, It took me several goes, to really find something that reflected what I wanted to create. Initially I was too focussed on just the facial expressions rather than the entire performance in context. This led to a really still and restricted sequence that wouldn’t have the clarity or readability needed for animation. After working with Ko, I was able to get a far better reference that had engaging movement and performance that would produce are far more effective and interesting animation overall.

My first step, which made a huge difference this time, especially in comparison to my Body Acting was that I started by drawing thumbnails. I think over the course of the Christmas break, I really forgot about the key processes of animation: that being Thumbnails – > Keys -> Timing -> Breakdown -> In-betweening. Skipping or accidentally reordering these vital stages meant that I was trapping my self in corners with animation that just fundamentally wouldn’t work, causing me to have to begin again and again. This is, even though the thumbnails themselves weren’t reflective of the final animation, I really took the time to work in order with caution and care and I believe this is evident in the final animation.

Once I was in the process of in-betweening the main animation, I found once again that I was using smear frames and speed as a crutch when I felt intimidated by fast movement. I was stuck on the idea that any quick movement had to use smears in order to be effective. Christina was instrumental in helping me understand that smears should only really be used in the most extreme movement and even then really minimally in order to have impact and not dilute the sequence. I think my main concern was that with out smears the animation would be slow and boring, but actually using them considerately in moderation maximises their effect and allows the animation to have flowing energetic movement without feeling odd or forced.

Looking at the below animations you can see the top V1, which is reliant on smears, in comparison to my final submission which moves far more naturally and successfully.

Below you can find some of the reference videos I used to create the butterfly animation. Whilst i didn’t want to spend too much time on the butterfly in comparison to the character, I did want to make sure that the character had enough react to and something that would work within the animation without being distracting. For this I used just three key frames in sequence to create a simple wing flap and then just relied on timing and placement in order to create flight.

What I didn’t anticipate was just how many frames it would take to finish the flight animation. Working on anything between 1s and 5s, the sequence ended up being around 550 frames, which made it almost as many as the full character animation which ended up being around 570 frames. This sheer number of frames as well as the complexity of the timing and synchronisation with the character movement, meant that I really relied on using image marks and notation for the first time. This proved incredibly useful and I went back and implemented it on my Body Acting files in order to really effectively and simply keep track of my different layers and sequences.

Body acting has been both an incredibly fun and simultaneously frustrating endeavour. Starting the project before Christmas, I was really looking forward to working on a full body animation that focussed heavily on performance, particularly one based on a specific character.

I chose Aron’s character, the mercenary turned gardener as I was immediately able to get a sense of how this character might move and then how to break expectations to make a light hearted and funny animation. In the one session before Christmas, I spent the afternoon simply learning about the character, how to draw him, the different elements of his design and exploring how he might move. After the first session I had a good overall concept of the design and had key frames I was ready to work on after the break.

After Christmas break, I came back into Body Acting with gusto, I was excited to work on the exercise and jumped straight into breakdowns and in-betweening. I found that this excitement was quite quickly dampened. Even though I had a clear idea of what I wanted, a video reference and a good understanding of the character design, I found it incredibly difficult to make him move in a way that didn’t feel stilted and forced. Lots of my early versions of the work feel very segmented and don’t flow well in terms of the full body motion.

After talking things through with Christina and Ko, I found that I was focussing too hard on trying to incorporate design instead of the movement and acting. I was so focussed on the character being “right” that it was never going to be. So with Christina’s guidance I decided to strip the design back to its very basic shapes, working from the same key frame ideas.

In this version, I stripped the design right back and tried my best to work in simple shapes, so that I wouldn’t be restricted by what I thought I should be doing. However I think that at this stage, I had been working at it for so long that I just felt blocked by the movement and that I couldn’t figure out how to make the movement feel possible let alone natural. So I sucked it up and decided to begin again. I went back and redrew my thumbnails, I redid my keys and my references and I started again.

This time I simplified the movement, I stripped the design back completely and I reformatted how I thought about breaking down the animation sequences. I realised that in the original sequence I wasn’t correctly breaking down the movement and really focussed on timing out the sequences, finding the balance and weight in the movement and working on efficient and effective movement.

This time I was far more successful in the exercise, I was able to understand the movement and execute them effectively, I was able to work actively with my references videos and I was able to exaggerate performance without relying on smear frames as a crutch. In hindsight i’m very glad I took the time to start this exercise over, it was a very useful lesson in not being precious about work and understanding when best to reframe your work both practically and mentally.