Week One:
Originally, I’ve been heavily inspired by 3D and 2D anime fights, most imparticular fights within Arcane. I love how fluid the animations are as well as the fight choreography. So I wanted to look more into arcane fights and try and replicate one of the fights.
When rewatching most of the fights, I decided that the Vi & Jayce fight scene was beautiful and thought I’d choose this one to recreate through pure referencing.
Since this was the fight I wanted to create I began by blocking out the environment and creating a rough outline of where all important objects would be so that I could map out the fight choreograph.
I found some models online of Vi from League of Legends with an Arcane style design, as well as a mock dummy rig to help be my enemies in this scenario. Because I couldn’t find a Jayce rig, I wanted to only copy the Vi-specific cut scenes.
After creating reference footage, I would use this as a template on how to approach the scene and use this as a guideline and storyboard going forward.
Week Two:
After messing around for a week, I suddenly decided to give up on this escapade I was eagerly excited for. I discovered the animation is a lot more in-depth than I expected, and it would be too unrealistic for what I had in mind and within the time frame I had.
So I, unfortunately, scrapped the idea and started from scratch with something that may be easier. Recently, I had been watching a lot of the anime Dragon Ball Super which reminded me of some earlier fights. I remembered the fights in 2D animations used a lot of repeated frames with motion blur and speed lines to help make the scene look more impactful than in actuality, and I thought it would be a great technique to learn. Not in the sense of learning shortcuts, but teaching me to not spend time animating things that were never going to be seen, which could help me develop the animation I’d like to as I’d spend more time working on what will be in the shot.
I remembered watching this fight many years ago, and a lot of Dragon Ball had stood out for me. With this in mind, I decided I would do a 30-second variation of this fight in my own form and personality to experiment and have fun, as fight scenes are a fun thing I’d love to develop as a skill with advanced mechanics.
Week Three:
First and foremost, I had to find a model of both the characters if I was going to do a Dragon Ball fight of these two, or it would be out of context, which is not what I wanted. The models I found and tested were:
https://agora.community/content/goku
https://milio-serrano.gumroad.com/l/RQWdd
The first major problem I saw was the quality of the texture and the controller quality. Both models did look and feel rushed in terms of animation controls, but the Kid Buu controllers were far more detailed and improved. However, the mismatch in quality creates an unnatural dynamic that would make it obvious and possibly distracting.
Although these were some immediate concerns, I found that it wasn’t necessarily important as I would only have to do an animation that would hopefully sell the fight rather than the detail of the models. But this made it quite hard to choose a background and create one.
Week Four:
I began animating my characters without a storyboard, as I would be keeping it simple in the sense of referring to the Goku vs Kid Buu fight video. I would start off with a dramatic pause, into a short clash, dramatic pause number two, clash number two, and a final scene. It seemed pretty simple, and I felt that I could create a similar dynamic to the referenced video.
I wanted to avoid rotoscoping as much as it was useful because I wanted to develop an understanding of timing without the need for footage (which is a big regret and naive approach).
By beginning my animation, I started by setting up my camera angles and scenes, so I would have a better understanding of how to compose the shots.
These would be the foundations of my animation, where I could work within these frames to build upon a fight scene.
Week Five:
The hardest part so far was to create a loopable fight scene that looked smooth. I was aware that unless I created a new pose for every attack, it would be noticeable, so my choice was to create a pose per keyframed pose, or to create a few and then loop them.
Because I was short on time, I ended up choosing the latter.
All of these poses were set 2 frames apart at 24 frames per second; I.e frames 4,6,8,10 would contain a new pose, with the in-between frames being animation clean-ups to prevent limbs from flying at 360-degree swings.
This would be the original clash animation which is a total of 22 frames of posing. This idea was to create a flashy impact that would let the viewer know there is an action pose occurring (a punch/kick/block), but fast enough to wear it seems like hundreds of attacks are being unleashed within an unspeakable amount of time.
Creating a cycled loop using infinity would help me save time and set up a scene in a way where I can get a close-up shot as well as a panned-out shot too for better fight dynamic and storytelling.
Week Six:
I wanted to create more personality within the animation by creating reactions, so I tried to do facial rigging of Kid Buu laughing or rather chuckling, but it came across as Buu talking or saying something – as to which I didn’t have any audio for, so looks incomplete and weird from a viewers standpoint.
Because a lot of my animation was achieved through looping animation setups and pre-made scene setups to my earlier prep, I was able to re-use existing scenes to help gain the same personality effect through short character development, such as my final scene.
I also decided to use no background, because it felt mismatching with the use of different textured quality of characters, and I was worried it would distract the viewers from the main focus of the animation. This being said, I kept the animation quite simple with lighting only using an Arnold Physical Sky and playing with the samples. This made rendering quite easy, which boosted my time left over.
Week Seven:
Because of my time leftover from last week, I am now able to spend more time on post-fx which will allow me to create an exciting dynamic. Starting off with anime speed lines. I feel that almost every anime uses speed lines to help illustrate the intensity of a scene.
On top of speed lines, I wanted to look for colored variations to help illustrate the tension and anger as well as the energy of the dramatic paused frames. For example:
The rest of the week would be spent editing layers and trimming seconds of certain scenes to help capture the intense fights by not dragging some scenes on. I felt snappy cuts would help keep the pace of everything up, considering I had about 15 scenes of 5-second clips. I felt that that would have been too long and dropped the appeal, forcing the viewer to search for the imperfections.
Given I had full control of how long I wanted the fight to be, I cut it quite short as 30 seconds felt more appropriate for the short fight scenes that I choreographed.
Summary:
Overall, for how little time I had before submission, I’m quite glad I could pump this out in time, as I felt I was heavily distracted with my previous projects which were dragging on due to poor planning. I feel that I made the same mistake with this project to tackle something that would be far too complex to do with my level of expertise, and knowledge of animation. Knowing my lack of experience, I still decided to challenge myself, but in all the wrong ways, such as not creating a storyboard because I wanted to create an improv animation.
Even with that said, I feel that I was able to create an engaging animation that had comedic value and was fun to create, and funny to watch, even if there were mistakes. I feel that this would be something most definitely I’d like to do again and do more in-depth as I had a great laugh watching my own ideas grow without the limitations of a storyboard, where it would have felt constricted and more serious.
Having the flexibility to change a fight spontaneously was the fun of this project, and being able to dictate the outcome is something I’d to focus on. Fight scenes have a great dynamic, and don’t need to be as intense and dramatic as a lot of animations can depict them to be.
The music was just an added benefit that gave homage to the anime, as sometimes theme music would play in the winning character’s favor, and I wanted to create something nostalgic at the same time.
I do not regret the final outcome, but I do regret my approach, and I wish for the next time I do a similar project, or any for that matter to take the planning stage more seriously by providing a mood board/storyboard. Even if I want to be creative or spontaneous it would provide the project stability that was required, which I believe I heavily lacked.