Link to the video (Dropbox)(Cannot be embed due to dropbox security)
Scene 1: My Childhood Struggles
Ever since I saw the movie IT(1990) as a kid, I was fascinated by its story, theme and characters in that movie. My fascination grew to movies I remember back in the 90s like Terminator, Batman, RoboCop, Jurassic Park, SAW and rated M movies. Hence I grew up where movies were dark in their storytelling and their characters. Thus I chose the dark tone/theme and I tried to include as many references/tributes to 90's movies in this memory.
Scene 2: My Journey as a Monster Hunter
I picked this memory due to my fond memories of playing difficult games while growing up. From PS1 RPGs to PS2 action games, the one game during that era that destroyed me the most is Monster Hunter Unite. The first scene the game introduce me is throwing me in the middle of a snowy mountain before being attacked by a giant monster. There were no tutorials to guide me thus I had to learn how everything works through trial and error. It's because of this game I learned conquering the game with just your wits and guts is the most satisfying feeling to have from a video game.
Scene 3: My dreams and my future
This 3rd scene is so different than the other scenes was because: 1) I want to see what I can do with my current skill-set. 2) I like to challenge myself very often. 3) I like to try out new and different things. As a kid, I always wanted to try creating something visually crazy like a Japanese Mecha in a vibrant style that shines.
Since this is the last scene I get to work on, I have put everything I learned since the start of the challenge into this last scene as a final hurrah. Starting this challenge as a programmer, I had to learn everything from 2D texturing to 3D modelling to Rigging to Sculpting to Animation. I also learned what's and how's basis of working in a 3D environment. For example, UV wraps, retopology, Animation IKs and Rigs and Shader creation to name a few. I also managed to dabble in how cinematography works and realized how camera work matter a lot more than I think thus I wish I can continue researching into it when I have the chance in the future.
To the other contestants if you're reading this, don't be afraid to do something out of your comfort zone and good luck!!!
Store Assets Tools: Unity's Post Processing, Shader Sandwich and UMotion
I drew a couple of quick concepts to start off based on my life experiences.
I chose 1, 3 and 5 as a bases for my scenes. I also quickly designed (or sketch as I call it) the character for my scenes. I did not want to do a human character because 1. it is too time-consuming 2. My skillset is limited 3. I want to do something different yet recognizable. Thus I gone with a humanoid robot with identifiable limbs and eyes.
It did not took me long to get a model out quickly(While making some minor changes) as well as going through UV Mapping/Texturing and finally Rigging it.
Next, I quickly set up the scene and learning the ins and outs of shaders. I used Shader Sandwich(which is available in the Unity Asset Store) to create the ground and it is self-exploratory(Kudos to its creator on simplifying shader process)
As seen in the Scene Setup image, I was testing out what Shader Sandwich can do. For this, I did Diffuse, Specular, Normal Mapping and Depth. The huge advantage I found from using Shader Sandwich is that the tool can output the texture as well as duplicating and modifying it into something else. Basically, inputting a brick texture and letting the tool generate Normals, AO and Speculars by itself. Thus reducing the back and forth between Unity and Photoshop and grants instant preview.
The props were made relatively quick. One thing I like to note is the container has transparency thanks to Shader Sandwich. The Wooden box was done the same way as the Ground model.
The glass ball itself is not anything complex. Its shader is made up of Diffuse and Specular with Transparency, Subdivide, Shells and Smooth onto the plain Sphere object. (Courtesy of Shader Sandwich)
I like to note that the balloon is very subtle in movement(Hard to spot but noticeable if looked closely). I did not animate that but instead scripted that due to time constraint and having more control over movement. As you can see above, the clothing the character wore was using a shader with tessellation effect to turn a simple plane mesh into that fur-like mesh. For the ghostly character, I made a shader with Transparency and Rim light around it to achieve that effect. I also positioned a light source inside the container since I found out it is possible to emit light from it.
I quickly modeled mountain using sculpt mode in Blender to achieve that mountain surface and textured it. I rushed it but it's enough for what it is.
While I already did decided on the scene according to my sketches, I felt it wasn't enough so I gone further and challenge myself on making something big instead. I drew and scanned a sketch and while I'm not a concept artist or a good artist generally at that, it is meant to be used as a base for the monster model which I can refine later on.
I also not a good 3D artist too hence in the 1st scene, I cheated a lot with shaders to hide the ugliness so to say. With this second scene, I can't find a good way to cheat so I had to set some time to learn sculpting. After going through video tutorials and manuals, I got a good grasp of it to be able to sculpt what I needed.
Then I made a custom shader for the monster with 2 layers: 1st layer has normal textures with some specular while the 2nd layer acts as the outer skin of the monster with Shell and Transparency along with its own overlay texture. I added in a particle system with a snow breathe texture I made quickly as shown in the image above.
Using the glass ball from the first scene, I added a mist effect to it by inputting another texture(with transparency) and manipulating the texture's offset in realtime. As for the other character, I reused the robot from the first scene and added in a sword model and parent it to the robot for posing. Lastly, I made a couple of refinements to the scene with the usual post processing tricks with Bloom, Anti-aliasing and Eye Adaption(to offset the brightness of some models).
I also added in a Flag model to help define the 'Goal' aspect of the memory. The Flag is made up of a long cylinder and a rectangle mesh with around 150 vertices and Unity's Cloth component.
On this scene, I completely threw out what I had in mind and decided to go crazy.
After modelling and rigging, here is the result:
First, I reused the glass ball from the 1st scene and the grassy shader to create the ground.
To color the scene, I used Color Grading to match the color theme and tone without changing any of the models thus reducing workload along with Bloom and Anti-Aliasing.
I used a particle system with a self-made texture for the purple sprites and a spot light to illustrate the character. The Bloom helps amplify the effects.
The character has two animations consisting of 'Intro' and 'Idle' were created by hand with an asset tool called 'UMotion' (Available in the Unity Asset Store). UMotion allows me to import my 3D model and recognizes its rig right away. Granted it was not perfect as the rig itself is a custom one as it doesn't follow a human rig and it has two extra limbs on its back. Fortunately they are easy to fix as UMotion has settings to correct all those rig issues. Next, I manually key-frames each poses then adjust the inbetweens. Sometimes, I had to resort to IK Rigs to handle some unnatural movement(UMotion made its IK and FK rigging setups intuitive to use). I would like to note the mirroring feature helps animating the wings a lot smoother and quicker.
The submission video I linked was done and edited by Adobe Premiere Pro as I do have some video editing background.
I learned a lot during this challenge. More so than I thought under almost a month of heavy reading and researching the basis, workflow, ideology and how to put into practice the best. While it wasn't in line of what I had from my early concept ideas, I'm more happy with the end results and the changes to them.
With the 1st scene, there's not much I think I can do other than refining the details to further its backstory a bit more. There was planned to be animated fireworks behind the camera, giving flashes of color lights but due to time constraints, I can't implement that in time unfortunately.
2nd scene, I wish I could have done more for the monster character. Like giving it an idle animation which gives it more life into it. The idle animation would have it moving its head to look around its surroundings before looking at the robot character. I also did planned to include other 3 robots with their own weapons to mimic the Monster Hunter games as closely but due to time constraints, I couldn't finish them in time.
3rd scene, I gave everything I know and have into this scene the most. At first, I did not intend to animate at all but after listening to a line that impacted me greatly ever since hearing it from the God of War(2018)'s director Cory Barlog, which I'll quote "Doubt kills Creativity more than Fear". That line made me go one step further and going crazy with ideas in my head. I know next to nothing about animation and with the limited time I have left, I decided to risk it and just ran with it. Thankfully, I think it paid off. Not only it looks visually appealing, it has life, movement and some personality. I also gained tremendous knowledge on 3D animation which will be useful in the future.
Note: If you have any questions, feel free to comment and I try my best to answer them.