Theoretical Text (Final version)

May 21, 2021

Theoretical Text (Final version)

This graduation thesis started from both main-stream Japanese TV anime show and internet music videos which influenced me every day. 3DCG is considered an assisting tool in traditional Japanese anime. Hand drawing is still the main production way for anime for the last century. However, in these few years, there are new anime that used 3DCG as their main medium while keeping the tradition Japanese hand drawn anime aesthetic. For example some commercial anime like, Land of the Lustrous, BEASTARS, BanG Dream!, RWBY, HELLO WORLD etc.… I have a great interest in this new field of 3DCG which called “Toon shader rendering” or “anime style CG”. I started as a completely amateur and leant all knowledge from the basic. As a result, I created a short music video which can highlight the strength of 3DCG and show the beauties of toon shader rendering.

In my point of view, “online learning resources is becoming richer and easier to access in modern internet community, thus amateur will have chance to approach professionals merely by self-learning”. I had experimented this argument through the production process of my final year project which focused on learning both toon shader rendering and designing a music video animation simultaneously. As a result, it worked partially.

At the very beginning of the project, I considered myself as an amateur animation director. I have a wrong idea before that is “a 3D artist does not need to learn 2D drawing as they can be separated in individual areas”. I felt afraid with my terrible drawing skill thus I have never tried to design any original human character before until I joined the industrial field trip on Japanese character design held by SCM. I designed my first original character at the end of the trip. Although it is just a rough draft with a shallow story background, it still helps me to step out from the comfort zone of “not able to draw”. It gave me a propulsion to draw more design maps and make a more ambitious project. Which is my graduation thesis animation.

This is my first time to create a fictional world with its own complete history, geography, ethnicity, aesthetic, architectures, characters, and their own stories. The original synopsis of the story was “a fight between a pair of non-human lovers.” I started the whole project with finishing a concept art of a marionette girl. But I could not fill this character with a rich story background. I did not know how she should fight and why should she fight in the animation. Then, I started to think of creating a history for the land in the animation world using the technique of automatic writing I learnt from SCM. From how a goodness created the land, how different ethnicities were born, how they get along, how they fight, how was the marionettes are created and how my main character become the only marionette that have a heart. Although these settings will not be shown directly in the animation, it helps so much on enriching the character and scenery designs.

The stage for characters to act had been prepared, the next step is to fill the character with their own unique experience and story. Title of my animation is called “beyond the red

string”. I wanted to make a strong linkage within “marionette”, “red string”, and “lovers”. At the beginning, I staged most of the scene including how characters enter the scene, how they fight and how the story end with a lethal sword stab. But I cannot fit the “red string” into the storyboard well. Stories of characters were being hollow, and I could not present the lover relationship of the two main characters well to my viewers. Luckily, the character design sheet given by my advisor helped me to overcome this difficulty.

The sheet forced me to imagine every aspect of a character, from board to narrow. From his/her dramatic need and his/ her attitude towards the world to his/her favorite music or sense of humor. I did not think this information will be contributed to my work until I filled in all the blankets. I found that it can helps me to understand and learn more about the characters that I designed. My fictional characters becoming a lifelike individual as if I know them in person. I can see their thoughts and predict their actions. I can generate story and memories for them as much as I can. These trivial information act as a good source to trigger my imagination. I can finally settle the design of the characters and the storyboards of the storytelling parts. I named characters with plants which have a flower meaning corresponded to the synopsis of story I created for them. For example, “Flanca” the marionette girl named from Casablanca lily meaning eternity beauty, unaffordable love and silence waiting. “Coleus” the cyborg named from coleus means despair love. “John” the human named from St. John's wort meanings animosity, grudge, and superstition.

I have never acknowledged the fact that off-screen settings are extremely essential. Before, I just tried to apply every visual element that I like on my characters to make them look pretty. But now, I know that this act will only make characters becoming hollow and boring. The key elements that completed my design are the concept of automatic writing that I learnt from school and resources from my project advisor. I may have designed a pretty character by learning from online drawing tutorial. But I will never create a lifelike character that fit my story well without the guide from provided from school.

Another obstacle of designing for me is to create the storyboards. Two main issues need to be solved: how to “draw” and what should I “draw”. In the previous class I took in SCM, the only way I could produce a storyboard is to draw stick figures with orthographic view. It looked hideous and difficult to be understanded by viewers. I did not have enough knowledge and practice in drawing. Such level of drawing skill will not be able to present all the dynamic action scenes in my animation clearly to viewers. I forced myself to learn and try to draw with correct perspective, however, it is extremely time consuming. It may cost me a whole day to draw one single storyboard. It needs time and practices to become a person who can draw well. I did not have that time to practice before the deadline of graduation thesis. So, I chose another way to finish the storyboards.

I watched the making-of video of “Fate/stay night: Heaven's Feel Movie” and found out how they used blocky dummy 3d model to simulate characters action in some battle scene. Originally, it was used as a rough draft for animator to draw on it. But I decided to use this kind of dummy model to form my storyboards directly. Now, I am not only drawing good storyboard in my mind, I also “find” good storyboards from the scene. Sometimes I feels like I am taking photo rather than drawing a storyboard. My imagination of composition was

limited by the anime I watched before. But now I can try out every camera angle, every focus length, and every composition to find the best camera shot I wanted and fit it into my storyboards. Although the speed of making storyboard in this way is still far slower than professional storyboard drawer, it is the best way for me to create storyboard at my current skill level.

I settled how to make storyboard. Then the next question is what content I should put inside the storyboards. Again, it is a difficult task I had never done before. For the non-battle shots, I cling on my memories about every anime I watched in the past and keep recording the interesting shot I found in the newest anime. I learnt from them, remixed them, and created my own storyboard. Two anime that inspired me the most are “attack on titan” and “land of lustrous”. One is 2D anime and one is 3D anime. Both has presented some top-level action scene in the whole Japanese anime industry. After I analyzed some shots of them frame by frame, I discovered there is a core trick to make every shot look more interesting. And the trick is “contrast”. If the previous shot is a long shot, then the next shot can be a closeup shot. If the last moment pan fast, then the next moment can be more stable. There are varies of contrasting elements keep appearing continuously in the shots to make them more vivid. From far to close, from fast to slow, from clam to excite, from scattered to dense, from curve to straight etc.… These concepts can always keep my shots away from boringness. So, I have applied them on my whole animation. Although it may be an obvious concept for professional directors, it is a precious treasure for me that I earned after a long journey.

The creation of those non-battling storyboard was being more conceptual while the creation of the battling storyboard was being more practical. At first, I have no idea for how to create a battling scene as I did not learn anything about fighting in my life. I started the production with finding some nice battle pose reference from illustration and anime. I analyzed battle scene from different anime. Then I discovered the conclusion: there are not much different between daily scene and battle scene. If you can create the keyframes, the animation will then exist in front of you. A fight scene can be easily done by just lining up random poses as keyframes. I found that no matter how different two pose are, there will always be a way to fill in the in-between. I created many intense poses for the characters included attack, defense, and dodge. Then, a series of smooth fighting action will be formed. Combined with the concept of contrasting, I finally can create storyboard of battling scene.

The last idea penetrated through my storyboard is how I managed silhouette of characters as simply shapes. I learnt the concept of silhouette first from SCM’s class “Character Animation”. I learnt that I should always keep the outline of the character clean and avoid too much self-overlapping for the character. Then, I heard from my classmates from the Japan field trip that a character’s poses will be pretty if they can be simplified into simple

Action reference sheets: https://sean-akaito.blogspot.com/2021/04/battling-pose.html

shapes. I tried to execute it and found it did give a great result. (I used triangle, quadrilateral, and D shape in most of the time. Some examples can be found on my blog.) Later in my pre-production period, I found a YouTube channel called “Doudou town” which taught illustration theory in a clear, simple and comedy way. He has already mentioned the concept of “contrasting” and “shapes” in his video and given examples of great artist use such technique in art history. Like Wassily Kandinsky and Alfons Mucha. Although I may have seen their art in my life before, I have never dig deep into their arts. I did not studied art in high school nor practiced drawing. I made a detour to reach these knowledges that are already found and summarized by these great artists.

I do not consider myself as a person who is good in “right brain stuff” such as intuition, abstract and conceptual theories. I feel pain and frustrated when I need to design and create things with no guideline like storyboard. There is no equation and formula to construct those storyboards. The only thing I can do to improve the quality of my storyboards is to watch as many great artworks as I can.

I succussed to finish the storyboards by self-learning only. However, I made a detour and waste lots of time on the path for discovering the concepts needed. I have an unbalanced preference on researching. I watch Japanese anime in most of the time and ignoring traditional arts in history. Ironically, these knowledges can be easily learnt from traditional arts. More time would be saved if I did not limit my research area in Japanese anime. As a result, I can say that an amateur can learn how to design storyboard by only self-learning. But he/she may go on a wrong road and make a detour is he/she cannot find a proper guide.

After finishing the storyboard, next process is to finish the model shader. I practiced 3D modelling a lot before and I have confident in this single area. So, it is not the focus of this thesis. The real issue I encountered with the 3D models was how to apply the Japanese anime style toon shader on them.

First, I need to find out what software I should use. The first software I tried is Arnold. It is a built-in software of Maya which have its own toon shader function. But after I tried for a while, I found it is not optimized for toon shader rendering. It is a buggy function that is still under development which means it is not suitable for production. Then, I did research on every great 3D anime made with toon shader. I found that all commercial anime used the

Examples: https://sean-akaito.blogspot.com/2021/04/theoretical-text-supplementary-image.html

same single renderer “Pencil +”. The industry is so small and undeveloped that I even do not have a choice on it. There is only one choice for me. (Some may use unity toon shader. But that is for the game side, not anime.) I used Pencil+ 4 for 3ds Max in my production as it has more advanced functions compared to pencil+ 4 in Maya. I read through the whole book of user Manual and explored all referencing file given by the manufacturer. I learnt how to operate the software; however, I have no knowledge in illustration coloring.

The mechanics of toon shader is far different from PBR shaders. In PBR shader, I will always get a realistic result no matter how I change the parameter. But in toon shader, everything needed to be controlled manually. From where the border of shadow should be located, how the gradient of bright faces and dark faces should be blended, what color should the shadow be, how to limit highlight and shadow in specific faces etc. Which mean I cannot rely on computer to calculate the light and shadows automatically for me anymore. I need to learn how to do coloring in 2D illustration and then convert it into parameters of the toon shader. I spent more than a month to practice the basic sketching and cell style coloring with the help of YouTube tutorial videos. I leant how to color different kinds of materials in cell style illustration like hair, silk, leather, metal, gem, liquid, and skin. In the process of learning coloring, I started to observed objects around me more. I tried to guess and figure out how light shines on different textures and how will light be blocked by different objects. The learning process brought me out of the comfort zone of over relying on procedural tricks. It brought me down to the foundation and reviewed the knowledge I should have known about images. Some said a good artist is also a good observer. After the process above, I think I have finally made my very first step towards a good artist.

Besides from character’s shader, I must also finish the shader of the scenery. I was decided to use RenderMan at first as I has learnt it in SCM and I am familiar with it. But when I am finding some tutorial that taught how to make Sci-fi scene. Most of them are using Blender but not Maya. At first, I decided to learn these skills and concept in Blender and then apply them onto RenderMan in Maya. But soon, I got impressed by the performance of Blender and migrated my whole production streamline of the scenery to Blender. It was a risky bet for me to change my production platform of scenery to Blender. As I am new to blender, I do not know how much time will cost me to learn a new software. Also, it will increase the complexity of my production process. I must process through three 3D software to get my final render result (3ds Max for toon shader, Maya for key animation, Blender for scene rendering). Data transfer between different 3D software will become a big problem for me if I choose Blender. After I balanced the pro and cons between going on a familiar path and take an adventure to a whole new world, I chose to take an adventure and use Blender. Blender has developed a new render engine called “Eevee” in the recent year. It can mimic the render result of PBR render engine while shorten the time by 10 times. Also, it can perform a real time shaded preview at the viewport even if there are some complicated elements like volume and particles. Furthermore, the community of Blender users is so generous that they will share all skills they know and resource they have freely to others.

More about render engine testing:

https://sean-akaito.blogspot.com/2021/04/theoretical-text-supplementary-image_27.html

The software improved the production efficiency of my work and the community taught me new knowledge to create. I learnt not only the software manipulating techniques from the community, but I also learnt the concept and process of thinking behind their works. Before, I relied on others’ result a lot when it comes to procedural shading. I used a lot of pre-made shader pack and texture map online to finish my work. The performance of my works is limited by how the original creators made those materials. But now, I know which and what node are needed to achieve the effect I wanted. I can have more control to my own work.

One core concept I learnt is to separate shaders into layers like I am drawing in illustration software. It gives more flexibility and variations to the shaders in contrast with using a single principled shader. Once again it let me realize that knowledge of 2D illustration and 3D rendering can shared. Someone should not limit him/herself in a single area of art. The more you know about art, the more you can create even if you just work in a single precise area. I have no regret on spending days to learn and jump to a new platform. I am also being lucky as I had found those great tutorials with high quality at the starting point. I was being guided on the right path to learn and grow which avoided me to make a detour and wasted time. It shows that online tutorial can provide sufficient knowledge and skill on a professional level if you can find one with high quality.

I have stepped far away from my comfort zone in the process of making this project. From an amateur who never learnt drawing to being able to design all the concept arts and colored all the models by self. From an amateur who can only present story with stick figure to being able to make well presented full-length animatic. From an amateur who only use default and pre-made shader to being able to create and invent my own shaders. With the help of online materials, I did learn, experience, and grow a lot.

But how far is the distance between me and the professionals? I would said I may be being close to them, but I have never reached them. I have seen some great graduation or individual animation project from local and Japan. For examples: “SKY FIRE – REDEMPTION” (Chu, 2013), “朱” (Mariyasu, 2019), “チルリ” (Kawasaki, 2011), “フミコの告白” (Ishida, 2009), “TOKYO COSMO” (Okada, 2015), “微睡みのヴェヴァラ” (syo5, 2019). I may have raised both my technique skills and conceptual knowledge to their level in the process of making this project. But the fact that I have not completed the animation separated me from those professionals. Nothing could be more important than finishing a project on time. I reflected on myself and found that this problem is resulted from two reason.

First, I wrongly estimated my working ability. Most of the skills I needed for the project is being learnt in the process of production. It cost me very long period to learnt and practice. However, I did not count these times into the working schedule when I was designing the goal of the project at the beginning. I made an over ambitious goal which is not suitable for an amateur like me at that time. My skills and ability at that time cannot afford this huge task I planned. As a result, the real production time is being extended to two or three more times than I thought at the beginning.

Second, I did not learn the importance of group working. Making animation is a tough task, it is rarely made by a single person. Most of the great example I watched are made by a

team of people. Even if the animation is mainly made by one person, he/she will still find some helpers do help he/she on some trivial tasks. Different people have different strength on different area. It will be more productive to place the right person on the right spot. Although the rule of this graduation thesis does not allow student to form group to finish their work. Student can still find somebody to help him/her and lower the workload. I was being stubborn on finishing all task by myself. If I had tried to ask help from my friends, production time of this project may have been shortened for a lot.

In conclusion, I have grown and learnt a lot in the process of making this project. Both knowledge from tradition art and skills in latest technology will help me to become a better artist in the future. The online world contributed a lot to this work. Great artwork can be found on web easily. The powerful searching algorithm can lead me to the specific type and style of artwork that I needed to learn from. Furthermore, software is being more user friendly and SNS are being more popular in the modern society. Creators can share their skills through posting tutorials online. These online resources are good enough to help someone to learn skills that the related industries needed. But the net only helps someone on the skills aspect, not the mental aspect. Being able to set a reasonable goal corresponding to one’s skill and the resource he/she have (such as time and manpower) is the most important task in animation production. It could not be taught by the net. It depends on how well someone know about him/herself and whether he/she has been guided correctly on the right path.

After the preproduction has done, I recognized that it is impossible to finish the animation before deadline if I keep clinging on the original animatic. My current skill and knowledge still cannot achieve the effects I planned in the original animatic. The path of learning will consume my time endlessly. So, I decided to stop. I make up my mind to abandon the original plan and finish a new version of animation in four weeks.

In the new version of animation, I aimed to make a short animation in a form of a teaser trailer. I designed the new animation based on skills and knowledge that I am familiar with as my advisor reminded me. Which is matching scene with music and compositing. The new animation is a big different comparing to the original one. The original one is full of dynamic camera works, full body motions and strong rhythm music. In contrast, the new version is built with subtle motions and calm music. In this version I tried to tell story through staging, composition, action rhythm, color, and symbols.

The story of the teaser can be separated in four phrases which can be distinguished with different color theme.

First story phrase is having a color theme of bright and dark. Coleus(left) and Casablanca lily(right) shown up and followed by two characters. The plants are symbols of the two characters. Coleus plant with the flower meaning of despair love represent the character Coleus. Casablanca lily with the flower meaning unaffordable love represent the character Flanca. The story is narrated from Flanca’s view in a spiritual world. So, the bright light shines on her as she is the main character. Coleus is a vague blurred memory fragment hidden in Flanca’s mind. So, he is absorbed in darkness, facing back, and going away from Flanca. Flanca acknowledged he may be a person who was important in her life. So, she reached out her hand and tried to catch this vague being. However, a mirror appeared in front of her and blocked her hand from reaching coleus.

Second story phrase is having a color theme of gold. Another Flanca appeared at the counter side of the mirror. Her pair of eyes are gold and limbs are tied with gold strings. Strings represents that she is being brainwashed and being controlled against her will. Gold colors represent John, who is the character that controlled Flanca with strings. John is also represented by St. John’s wort, which has the flower meaning of animosity, grudge, and superstition.

Gold-eyed Flanca represent her mind being controlled. Red-eyed Flanca represent her original pure mind. The composition of the shot presents that the controlled mind of Flanca herself become an obstacle for her original mind to reach the memories of Coleus. While her leftover original mind is also being an obstacle that guarded herself from fully controlled by John. The two-mind then wrestled with each other. At last, the original mind lost and faded out.

Third story phrase is having a color theme of blue. Blue represents the subconscious and deep mind of Flanca. Flanca’s mind was abandoned to the subconscious. She dropped and dived into her own subconscious. She saw different memory fragments between her and coleus. In the memories, Coleus saved Flanca from a war field and treated her well. Coleus gave her a Casablanca lily headwear and an umbrella as gift. They get along well at first. But later war started again. Coleus escaped with Flanca from the war for a long time. At last, Coleus found an ancient artifact with a mysterious energy that can end the war. Unfortunate, they were in John’s crosshairs. John caught them with his army, snatched the artifact, separated Coleus and Flanca apart. And finally, John brainwashed Flanca and controlled her.

The tower in the scene is a symbol of “The Tower” in Tarot. Which represents destruction, accident, misfortune, and lovers having retribution that correspond to Flanca and Coleus’s background story.

After remembering the old memories with Coleus. Flanca clung on the object that have the strongest connection to Coleus, which is the umbrella that given by Coleus and Flanca always keep it by side, and leaved her subconscious.

Fourth story phrase is having a color theme of magenta. Magenta represents the spirit of Coleus. Flanca leaved her subconscious and stepped into the spirit of Coleus. She was surrounded by the soul of Coleus. Then, a fragment of Coleus’s mind shown up. He gave Flanca a hug. The hug from Coleus strengthened Flanca’s mind and destroyed the connection string. Flanca was freed from John’s control. But this Coleus was merely a fragment of his leftover spirit. Coleus in the real world has left Flanca for a long time. So, this spirit of Coleus disappeared soon after helping Flanca to destroy the strings. As Flanca has got rid of John control, color of her eyes recovered to red from gold. She knew that Coleus in real world has left her long time ago. She knew that she should not look back and she should step forward. So, she stepped forward and destroyed the St. John’s wort in her mind. At last, she walked away from her past and started her new journey with her precious umbrella that received from Coleus.

There are three main aspects I wanted to present to the audiences from this animation. First aspect is the color theme. I tried to produce different scenes with a single outstanding theme color. For example, gold, blue and magenta. As high a concentration of single color can leave a strong impression in audience mind so that they will remember the animation more. It also does contribution to the storytelling as it can separate different phrase of the animation story clearly. The most successful part to me is phrase three with blue color. Different saturations and levels of blue are being harmonious and resulted a glamorous scene. However, I am not satisfied with the scene in first and last phrases. The color theme of first phrase is bright and dark. I think it is not clear and outstanding enough. Color of the scene still look a bit messy. Besides, the last does not distinguish from others. The color theme of it is blue which has repeated the color theme in phrase three.

This problem is resulted from production time. Although I abandoned the old animation plan, I keeps the old production process which is to use multiple 3D and video compositing software to finish my work. Scenes with good looking color theme is produced through works in Maya + 3ds Max + Blender + Nuke + After effects. Each software serves their own purpose. The current result can be only achieved through using all of them. The work process is complicated, it takes time to study ways to create such scene. After counting on the schedule, I found that I do not have time to make scenes from all phrases into different individual single-colored theme. So, I made a choice to forgo on making the first and last phrase to be perfect. The first phrase is produced with three but not five software. And the last scene is produced with existing materials. With these two methods, I reduced the time needed for learning new skills and try and error. I can finally finish my work on time. If further improvements can be made, I would change the scene in the first phrase to white color theme and the last scene to iridescent color theme.

The second aspect I wanted to present to the audiences is the symbolic and abstract storytelling elements. I have implied the story behind Flanca through staging, composition, color, symbols etc. While not letting characters to act out the story in an obvious way or tells the story through voice over and characters’ conversations. As I wanted the story to become more reserved and more obscured. In this way, audiences will have more space to think and imagine about the story. Each audience will get a different experience after watching the animation.

This aspect of result is resulted from my carefully designed preproduction work for story and characters. With the helps of preproduction materials given by advisor and the long time that I worked on preproduction, I have designed detailed character background story and world setting. I know the relationship between these fictional characters clearly. I know how they react to each other, how the world influenced them and how they see the world. I know them well so I can present their story in any type of animation. From the solid action animation in the old plan to the abstract symbolic animation that I produced now.

Nevertheless, there is still having a large space for improvement. One big issue occurred is that the screen time of every abstract/symbol/cue for explaining the story is too short. Shots change fast in this animation. Each shot mostly last for one to three second only. Audiences may not have enough time to understand the meaning behind these composition and symbols. One solution for it will be replace the background music. I design scene and shots based on the rhythm of the music. If I can find an alternate music with longer length, I can extend the length of shots a bit more. Which can make audiences understand the story more. Also, adding text may be another solution to solve the problem. I consider text as another kind of symbols. It will not be contradicted to my idea of making the story to be reserved and it will even be a great help for explaining the story. But I am not familiar with typography. Base on the principle that I should stick on the skills that I am familiar with to finish the work on time. I decided not to add text to this animation.

The third aspect I wanted to present to the audiences is how the shots synchronized with the music. I designed each shot and action to be continuous even if the scene and story phrases has changed. Each single shot is linked to the previous shot. The animation is like a smooth linear journey for Flanca to dive down deep into her heart. And this effect is resulted from designing storyboard along music. I sticked on two principles when I choose the background music. One is to find music that have distinguishable pure tones. In this case, are the high pitch piano tones. I treated these tones as sound effects. I snapped keyframes to these tones such that the actions will be synchronized with the music. Another principle is to find music with clear main rhythm. Rhythm in rock or heavy metal may be too complexed as multiple rhythm are being mixed. While in this animation, I used a calm piano music with song of fictional language. The piano gave me a simply and clean rhythm which I can extract the emotion inside it and convert it into action and animation then create the storyboards. Also, the song was sung in fictional language with no practical meaning. The song weighted more on the abstract emotion side then the practical word meaning side. Which become an extra material that helped me to develop action and emotions in the storyboards further. With knowing every emotion in the music clearly, I can figure out ways to connect them and form a continuous flow of animation.

But there are still some issues happened in the process of managing the animation flow. First issue is that the pace of the animation is not stable. Sometimes it is too quick and sometimes it is too slow. Some parts in the animation did not perform the continuous and smooth pace that I wanted. And part of it is caused by technical problems. As some scenes are heavy for computer to calculate, preview result of the shot will be played in low frame rate. I can not get the direct and intuitive preview from the software. The preview result is different from the final rendered result. So, sometimes I set keyframe and values by guessing. I hope that I can get the correct pace of animation by luck. But it does not work. So, the situation of pace in shots being non-consistent happened. The solution for it will be using the traditional ways to set keyframes and in-betweens. Which means I should gather a stopwatch and a stack of X-sheet to plan every frame of the animation carefully. But not done it by guessing and feeling.

Second issue is that the motion direction sometimes is not consistent. I aimed to make the motions in a smooth linear flow. The motion direction in the current frame should be an extend of the previous motion from the previous frame, however, I broke this rule in some shots. For example, I added an anticipation to Flanca’s hand when she is reaching Coleus. It makes the shot look more natural by individual. But when it is combined with the previous shot, the anticipation is going to a counter direction compared with the previous motion. Which is breaking the smooth animation flow and gave a bad result.

This is caused by the roughly planned storyboard. In the storyboards I only done the staging part. I only located different characters and objects in the shots. I did not have a clear idea for the detail motions. I designed the motion in the production period but not in the preproduction. Which caused a problem that sometimes I only focused on the current shot and forgot the motion design in the previous shot in yesterday. The solution to it is simple. It can be solved by adding more frames of storyboards in the animatic and more instruction to indicate all detail that should be presented in the final product.

To conclude the production part, I made a great change in the final production. I abandoned the whole original animatic and created a new one. Which can made me to stick on skills and knowledge that I am familiar with. I learnt that I should make up my mind to forgo a plan that is not feasible to finish. I should design a plan and schedule by referencing my current level of skills. I have produced the final product successfully after I remake my plan. I produced an animation that tell story with composition and abstract details. I am stratified with most parts in this animation. But there are still some improvements that can be done. All these improvements are related to preproduction. If I have made a feasible plan at the very beginning but not at the last four weeks, I will have more time to focus on the preproduction specifically for the current version of animation. Those issues that I am not satisfied with can be solved. Which reminds me again that making a feasible production plan is the most important thing in animation production. And it is also what I learnt the most through the long period of making this graduation project.

Reference (individual animations):

Chu, Y. Z. (2013). SKY FIRE – REDEMPTION. https://vimeo.com/68701770

Ishida, H. (2009). フミコの告白. https://www.youtube.com/watch?v=0QqT1P4VO30

Kawasaki, K. (2011). チルリ. https://www.youtube.com/watch?v=_VK2il0LnOY

Mariyasu. (2019). 朱. https://www.youtube.com/watch?v=BJa2g4bLBq4

Okada, T. (2015). TOKYO COSMO. https://www.youtube.com/watch?v=3SEXokSxZBI

syo5. (2019). 微睡みのヴェヴァラ. https://www.youtube.com/watch?v=z4lIbrUPcag

Reference (commercial animations):

Graphinica. (2019). HELLO WORLD.

Orange Co., Ltd.. (2017). Land of the Lustrous.

Orange Co., Ltd.. (2017). BEASTARS.

Rooster Teeth. (2013). RWBY.