Pipeline Research and Planning

For this concluding term, I need to be efficient in planning the final steps so that I can finish in the time I have left. Starting from May 30th I have until the 19th of August to complete everything I can. So that's only 2 months and 2 weeks.

I'm going to start by having a better sense of the overall pipeline steps I have left so this is some brief research into the stages and options I have to do this term to complete the film.

Much like that one pipeline comic that seems to be used a lot when discussing the pipeline, for me, it is useful to break down the stages into the smaller steps that go into completing them to have a better understanding of things and essentially a to-do list of everything that needs to be done.

Texturing

• Unwrapping

• UV texture mapping

• Putting the section into different shaders

• Baking in Substance Painter

• The Painting Process

• Assigning new Materials on the Maya Model

Rigging

• Installing Advanced Skeleton

• Weight/skinning

• Face emotion checking

• Possible use of Blend Shapes -increased control of the face

Animation

• Camera Set up

• Camera Animation

• Character Key Posing

3D VFX

• MASH physics

• Replicater

Lighting

• Toon Shader Has Specific Settings

• Rim Lighting

• Stylised Lighting

• Arnold Documentation https://docs.arnoldrenderer.com/display/A5AFMUG/Toon

Rendering and Toon-shader

• AiToon - Using Note editor use both texture and toon shader

• Possible use of Ambient Occlusion

Compositing

• 2D Matte Backgrounds and Foregrounds

• Other 2D elements

• Dust

• More objects other than boxes

Sound Design

• Outsourcing

• Finding references

• Giving sound directions

Diving into the 3D World

My plan this term is to get as many of the 3D elements for my Masters Film built and modelled as possible!

What will be 3D and What will be 2D?

From the beginning, I have been planning to explore how both 2D and 3D are used together to create more exciting visuals. This is also a way for me to learn 3D as it is something I have never dived into before. To break down what will and won't be 3D amongst my scenes will be helpful in planning what needs to be learnt for where!

I am very aware of how vast the world of 3D is, therefore my plan has always been to find out exactly what I need to use to complete my film and try not to get too distracted by all the different ways of doing things.

What will be 3D?

• The Character

• For scenes where the camera is moving around the environment.

• This will make the process more efficient as I will not have to animate things in heavy perspective by hand

• For scenes with specific camera moves e.g. the dolly zoom

• The Environment

• Mostly for scenes that are entirely boxes, when there are more detailed backgrounds with fewer boxes that might be a 2D painted background.

• There will be a scene of boxes falling that will need to be simulated in 3D

• Most of the scenes that require heavy perspective with camera movement

• Making uses of programmes within Maya such as MASH

What will be 2D?

• The flashback scene will be entirely 2D to show a nostalgic feel

• This will involve two 2D characters and a 2D background

• Effects of dust particles (to be researched)

• Some character animation

• Large scale backgrounds may be painted

How will I make the combination work?

Once there are finished models to work with I will start creating render style tests. The aim is for the final rendering to have a painterly, 2D like finish.

This is a common style known as toon-shading. This hybrid 2D and 3D animation style was the subject of my Research Poster in my last term. From that, I studied the ways it is being utilised in recent media and how there may be a 'renaissance' of sorts when it comes to using 2D aesthetics in feature animations. I also looked at the reactions people have to each of the styles, as my Masters' film will be one full of sentiment this was an important aspect for me.

References:

Arcane (2021-present) is a mostly 3D animated TV show, however their 2D VFXs, their Matte painted backgrounds and their painterly style, have greatly influenced my plans for my project.

The Character Model

Zbrush

I started by using knowledge from the last term to start constructing my character in Zbrush. I used modes that were displayed in Sean Yu's tutorials as well as some that I found through personal research. I found the tool 'Zsphares' particularly helpful to block out the shape of the character and experiment with the certain posture I am going for.( More on this in my Character Design blog.) Following a tutorial from Flipped Normals, I was able to start to see the character come together!

After this blocking out was done I then proceed to add in details. For some parts of the model, I created separate subtools to form the different aspects such as the hair, eyes, clothing and hands. I also made sure to change my character's pose to a T-pose to make it easier for the rigging stage.

Within this sculpting process, I learned a lot about the different kinds of tools and brushes available that can help with each stage. For example there are a series of particular brushes solely for working with fabrics, which was helpful when modeling the clothing for more character. I didn't want anything too complicated for the character's outfit but being able to at least give the model some fabric details such as folds and crinkles made it still look the natural way it needed to.

Maya

Retopology

The next step I have now is to move the model into Maya to complete the Retopology. This involved constructing the low res quad mesh that will cover the model and allow for the animation to be smoother. To make the process of hand constructing all the quads over this model I started out by only filling in half the character and then mirroring the mesh onto the other half.

There were some problems to solve everywhere within this process, but what really helped me was finding references to base the Quad pattern on. The main place I had for finding these was Pinterest and I was able to collect a few clear examples of Retopology for hands, faces and converting high loop numbers into lower numbers.

I also found that the closer I got into the mesh I had to turn off the 'Live' option on the original decimated mesh from Zbrush, this is the setting that makes your quads stick to the model and hold its shape. By turning its off I was able to more freely retopolise the nooks and crannies of the character. This was particularly helpful in parts such as the ankle, tin between the figures and when creating a mouth bag.

The mouth bag involved forming an inside of the mouth so that if the character was ever to open their mouth there would be geometry to support that animation.

Rigging

For rigging my character I wanted to give my full focus to the process to properly understand how to get a working rig and unfortunately, due to the length of time to complete the model and retopology I couldn't commit to the large scale tutorials needed for rigging before the deadline. This is also one of the sacrifices of directing the group project that I found I had to make to keep on top of that module. However, I believe next term will be a lot less taxing without having to direct another project and I intend to fully commit to rigging my character in full ready for animation and final production.

Other Issues with 3D

Another setback I have come up against is that my home PC is not capable of running any 3D programmes. I have tried to instal the student licensed Maya with no luck with things functioning normally. In the first term, I solved this by remote desktop working, however for this term the lag this created was far too much of a hindrance, particularly to Zbrush Sculpting. So I have been commuting in to work in the 3D Lab as much as possible. This has certainly set me back in relation to those who can access it at home, however, I have still managed a fair amount whilst learning the whole thing from scratch.

The Environment

The main goal with the environments this term is to see methods I could use to block out the scenes that will be in 3D. To start with I have been building a basic wall of boxes, using duplicate tools I have been able to make this process quite efficient. I have been adjusting the boxes to show a variety of shapes and sizes and positioning them so that they are precariously stacked as if at any moment they might come crumbling down.

MASH

Using this stack of boxes I then ventured into the basics of MASH. This was a very fun tool to use and I found it has really opened my eyes to the possibilities of what I can do with my environments. By duplicating my stack of boxes in instances it allows me to create a large amount of detail within it slowing down the programme and rendering time.

Scenes from my animatic that will work with MASH

This particular scene will work well with the physics options:

MASH Images

Lighting Tests

I have also briefly experimented with lighting on these boxes. As the majority of my shots in my film will be almost monochromatic, the simple lighting options in Maya actually provided a very close interpretation of my animation lighting and hues.

I dimmed the Skydome light and added a directional light that I experimented with turning slightly purple to match my animatic style. I then played with the directional light in different angles.

The Going Live Module: The Group Project and Industry Brief

This module 'Going Live', is our industry module where we are given a brief from a company in the industry of animation. The company working with our year is Wild Child Animations, a studio based in Stirling, Scotland. They are a BAFTA Scotland award-winning company that is steadily growing after the joining of two different studios in 2020 founded the company. I am particularly interested that they produce both 2D and 3D work as I'm coming from a 2D background, learning 3D.

The brief given to us was to make a short ident for the proposed educational branch of their company, Wildling Studios, with a limit of 30 seconds. An ident is a short and sweet introduction to a brand, usually a studio or something that works with creative media, the best example I thought about when considering idents were the old Channel 4 ones that would display large '4's in a scene.

Above is part of the concept I presented to Wild Child Animation in the pitch stage.

The concept that I discussed and developed with ideas from Kelly Clayton and Luke Coventry, was the idea of a POV shot exploring the inside of a maze that eventually leads to an opening that reveals the words: "Wildling Studios". The maze is formed of different tools and icons of typical animation software user interface (UI), these references help suggest the maze is the journey an early career animator may face. There are so many different options to go down when you are starting your career in the industry, the solving of this maze to reveal "Wildling Studios" is the light at the end of the tunnel to help early-career animators on that journey.

To better illustrate this concept I drew out an animatic which we also presented to Wild Child.

Relating to my Masters Film

This coincidently was informative when designing shots for my Advanced Production project, the way the corridors in this maze grow dauntingly confusing as well as the useful practice with the changing perspectives with a moving character's view, were qualities I will be using within the short film.

Stills from Box Project WIP animatic

Indent Pitch Feedback

However, it is this particular daunting and confining nature of the animatic that Wild Child had the most feedback about when presented with it. Rightly so, they commented that being within the walls of this maze was a bit too confined and conflicting for the idea of the ident. Idents are to represent the brand, usually with a few fun and interesting visuals to help the brand stand out, so this feedback was definitely understandable.

Despite this, out of the 6 ident concepts that were pitched to Wild Child, the Maze concept was one of two that were selected to be completed this term!!

Adapting the concept

It felt amazing that Wild Child could see the potential of this concept, even with their points about the confined POV of the original animatic. So, having been given the opportunity to respond to that feedback, I and my team had to try and adapt whilst still maintaining the original concepts that Wild Child was drawn to.

The main concepts to maintain in the new plan are:

• The idea of solving a maze like a puzzle

• Wild Child specifically said that they liked some of the references we showed in the pitch mood board that displayed a maze from more of a birdseye point of view.

• this lead me to think of maze-like puzzles that could be solved like a jigsaw or a Rubik's cube

• The various animation software, UI references

• They liked the idea of the UI being used creatively.

• Making these otherwise very industrial, lifeless, functional icons and imagery into something that shows the results of their function.

• We also wanted to get in the idea of growth

• something connecting the process to growing and learning

• The final thing must show the brand title Wildling Studios as the reveal

From this, the team have been looking into the idea of having this puzzle, similar to a Rubik cube, suspended in space, and being solved in front of the viewer. As it is solved we see the letters of Wilding Studios line up across the puzzle to reveal this is the answer. To display the letters and show reference to the original pitch, and the idea of the animation industry as a 'maze', we kept the pattern of a maze fixed on the outside of the puzzle.

The Cube

The way we would show this puzzle went through several iterations from a sphere to a small world of mazes, but between us, we decided that a puzzle inspired by a Rubik cube would be the most recognisable form to take.

However, we did some concepts and a storyboard of the solving of a standard Rubik cube and I couldn't help but think that seeing it in its simple cube shape the entire time it is being solved was a bit visually uninteresting. I started to wonder how we could make it look more unique and satisfying as a whole animation. Something important to remember however was that it had to still be physically possible to work the best in the viewers' eyes. This is why I turned to research different shapes and types of Rubiks cubes as I knew these would always be solvable and make physical sense.

I first looked into a Rectangular Rubik which would take on different right angles and different blocky angles in its solving, it would then show 'Wildings Studios' quite nicely on its long side when finished. However, this was when I also came across the 'Mirror Cube'.

I thought the way it looks when it is in the process of being solved was beautiful in its own way, The way its final form takes it back to a simple cube regardless of how different it looked whilst being solved, to me, felt like the process of any creative project. No matter the shapes it takes in the process, the final thing can be just as you meant it to in the end, or better! This, I feel also created for a much more satisfying finish visually as well as meaningfully.

Concept exploration drawings by me

The UI

When deciding how to incorporate the Animation UI tools within the indent we tested a few different ideas. The original was to have the UI appear first and form the maze itself with other references sprinkled in and around the maze. however, with this idea, the UI would start off in a 2D style and we had trouble planning out a smooth and seamless transition from 2D to the 3D model of the maze cube.

To simplify this, we will just have references to UI icons on the cube itself. This also went through several iterations and techniques. There were multiple routes we could have gone down. There were tests to show the UI has an animated texture, with aspects moving and glowing across the cube. We also tested having the UI within the 3D geometry of the cube, this looked good but created questions about how the cube sections would seamlessly move against each other. In the end, we are going with a texture constructed from scratch by Kelly Clayton in Substance Designer.

There was also a question or where the UI will go on the Maze Cube. Would it be hidden details within the sections of the maze? Would it be floating around in space? Looking at the concepts we are developing for the unsolved look of the Maze Cube, there are large sections of the cube that will ultimately be facing inwards but whilst being solved these sections would be visible. In concepts, we had been giving the outside facing points of the cube the bright brand colours from Wild Child's brand guidelines, and we had been leaving these inside facing sections grey. By creating some concept art to illustrate the idea I pitched t the group that this is where the UI elements could be displayed. Like a lot of points within this project, I feel this idea worked on multiple levels. The reference we are making to the animation UI is to display the tools we as animators are so familiar with; this, as well as being the connections this ident is making to the animation industry, is displaying what is on the inside of a production, the 'behind the scenes' that isn't displayed on the final product.

A concept a drew from our 3D animatic to help show where the details will be

The Vines and Growth of Nature

Something we were keen on getting within the scope of the project was the connection back to 'growth' and nature. This was particularly fuelled by positive feedback on those aspects from Wild Child! It made sense to try and relate it back to the 'Wildlings' and 'Wild' life nature of their brand.

The main idea we were pursuing for a while was vines growing around the cube and through the mazes. We also thought to have other styles of nature popping up. However, something that became apparent in another talk with Wild Child was that we had to be careful not to have the nature looking too much like it was overgrown and unkept, like moss and vines growing on an old abandoned building. To rethink this we dialled back on the number of vines growing around the cube and keep it only in the areas of the Letters. This helped cut back on the processing time needed on the animation as well so was useful.

Despite this, we still wanted to figure out a way to show a big burst of growth to be the result of completing the Maze Cube and finding Wilding Studios. So we moved the majority of the nature growth to burst out at the end within the lettering of Wildling Studios. This quick burst, rather than slow, helps to show the energetic, childlike nature of this company and not an old ancient one.

To help clarify the concept to the team before production on the main 3D model began, I used stills from the simplified animatic and filled in the FXs that the cube will eventually have.

We are in production with the final cube model and animations being put into place and looking forward to showing Wild Child Animations in person soon.