ANIMATED SHORT: BRAVERY

For our final project of Semester One we were tasked with creating an animated short. The theme of this task was Bravery. I was on a team with Travis Morrow, Matthias Jaques, Hristo Petrov and Sudipta Patra. We began as soon as possible with sharing our ideas. In summary, most of our ideas involved quite simple concepts such as a character taking a leap of faith of sorts etc. Su had suggested the idea of a Samurai preparing for battle and we all were on board quickly. It became clear that we were to convey an implied sense of bravery throughout our film, rather than an explicit manifestation of the theme. We felt that, in doing so, we could focus on using visual storytelling through composition, colour, sound and editing.

This brought us to creating a storyboard. We all agreed that it would be fitting to have the majority of the film to centre upon the samurai getting ready for battle, with a final pay-off with the end shot. The pacing was to begin slowly and gradually speed up to create a feeling of tension. It was at this stage that we decided to name the film 'Ikigai', meaning; "a reason for being". We all drafted the storyboard together to remain clear on where our story was going. I then created the final version that we could all reference for the Animatic stage.

We translated this storyboard into an animatic quite quickly and we were all happy with the result. Our main aim was to focus on composition and timing at this stage of the process so that we could iron out any flaws with editing/timing before we began modelling. Below is a video of the first animatic.

We received some feedback from both Alec and Gerard to place more emphasis on some of the cuts in the editing so as to give more impact  as the speed builds up. They suggested we also provide some more backstory to the character to add in more depth. So, we began to think about utilizing the mural behind our character to show a tragic history. As a group, I think we took on the advice and tried to use it to  better our overall idea.

After this, we split into sub-teams; a character team and an environment team. Travis, Matthias and I decided to focus on designing the environment, while Su and Hristo took on the character design. I started to design some concept art of the landscape and surrounding environment. I tried to come up with a few colour schemes and some lighting ideas. In terms of style and colour we wanted to reference the likes of 'Okami' and 'Samurai Jack'. 

We decided that we would use Allegorithmic's Substance programs to texture our scene. I first began to look into Substance Designer to texture the flooring and walls of the samurai's temple. It was challenging at first as I wasn't accustomed to the software's interface. However, gradually it became clear that the node-based system was quite straightforward. By following some of the tutorials on Allegorithmic's website and YouTube, I was able to get to grips with it and create tileable textures such as wooden flooring, concrete tiles and a Japanese paper wall.

I then took to ZBrush to play around with sculpting. I wanted to experiment with different styles of Japanese sculpture. The 'Komainu' is a lion-like creature which guards the entrance of many Japanese Shinto shrines. 

We decided that we would aim to use Unity as our render engine early on in the process. This meant that we would have to change the way that we thought about the modelling process. We would have to assure that most of our models retained a low poly-count if they were to hold up in a real-time game engine like Unity. We were inspired to use the engine after Hristo showed us a few Unity demos such as 'The Blacksmith' and 'Adam'. We were aware that this would be a rather steep learning-curve for us but we felt that the advantages and end product would be worthwhile. It would mean that we could almost live-preview our rendering as we went along the process and edit it as we saw fit. I thought that this would be extremely efficient to use as a team.

This brought with it a few issues, however. It would mean that we would have to figure out how to maintain high-quality models, but with a low poly count. Substance Painter was our answer in this situation. It was very useful for us to be able to craft normal and height maps so easily, which sped up the texturing process immensely. I also experimented with baking high-poly to low-poly with xNormal. The results turned out well after retopology and baking. I was able to reduce a 400,000+ poly rock model to a 301 polygon model with little difference. I then added more texture and Japanese engravings for the final piece.

Similarly to Substance Designer, learning Unity was definitely a challenge. It required getting used to a whole new software. However, it was actually very similar to Maya once I figured out how to navigate it. We had to decide how to create a terrain to use in Unity to set the scene. Unity has a built-in terrain system that allows you to mould a landscape. At first I used World Machine, a program that allows you to craft terrains procedurally. This did work quite well as a terrain maker and looked nice and mountainous. However, we discovered that this would not actually work out with our Maya to Unity workflow as a Unity native terrain is not an obj/fbx or any component transferrable to Maya. What we needed was a terrain that could be used in both Maya and Unity. Therefore, I made a terrain in Maya by using a Mt. Fuji height map and moulding it to our liking. I then textured this using Substance Painter. I attempted to make it look as natural as possible and utilized the various brushes to generate height and detail.

Original Unity Terrain

Final Terrain

We then began to assemble the scene using Travis and Matthias' assets. I then added post-processing effects in Unity such as motion blur, depth of field, colour grading, etc. We played around with different times of day and sunlight to give ourselves options when selecting colour schemes. We all seemed to gravitate towards a reddish looking sky as if the sun is just rising. We felt that this was both aesthetically pleasing and symbolic of the Japanese Rising Sun.

Below are some lighting tests in Maya

Once the we were happy with the scene, Matthias and I began to animate Hristo and Su's character model/rig. We split the shots in half and spent most of the process together animating side by side so as to give instant feedback and critique. This was quite an effective method of animating it and quickly ironed out any problems.

My shots included:

 FINAL REFLECTION

Looking back on this 6-week project, I'm very pleased with the outcome our team achieved. We wanted to challenge ourselves by learning new software and applications. At first it seemed daunting to have to learn Unity, Substance Painter, etc., however with the help and reassurance of my teammates we were able to overcome any problems we encountered. We all had faith in the story and that we would be ale to deliver an adequate result. We worked efficiently together and always maintained good communication which I believe helped the project to run smoothly. At several stages throughout the process we sought feedback from both lecturers and fellow students to gain some outside perspective and ensure that the project wasn't going down the wrong path. I felt that our attempts were successful in the end. It was rewarding to see our efforts as a team come together and we were all very satisfied with the end product.