Pimpaka Lertjittikun
Eating Deity
UNIVERSITY THESIS
2022 ~ 2023 ( 6 ~ 7 months )
This is my final project before graduating. I decided to make a 2.5D-ish art-style horror game with the Theme of Thai temple. I wanted to experiment with the whole pipeline of making games other than art, and this is the perfect opportunity.
Game mockup visual
DEVELOPMENT PROCESS / TOOLS
Initially, I wanted to make a Fez puzzle-like game with a 2.5D art style. Starting off, I needed to set a scope, which I've failed since I thought I could make one indie game. Then, after scrabbling through programming and game design, I found that scrabbling through art stuff is a lot easier. The process is a mess, but I've got to experiment with a lot of stuff that I've never thought I could do.
Tools
-
Unity
-
Created a character controller with a camera controller
-
Create shader value controller
-
Cinemachine
-
Created a camera that follows a player
-
-
Shader Graph
-
Visual Effect Graph
-
Animation
-
​Create sprite animations
-
-
Photoshop
-
Create pixelated textures & animations
-
Create concept art
-
Create other art assets
-
-
Blender
-
Create 3D environment assets
-
-
CONCEPT ART & DESIGN
Concept Art time! At first, I thought this would be as straightforward as it is: established mood & tone + final mock-up visual so that I can aim for something while working in the engine. Instead, there needs to be some amount of game design & narrative/story here required before concept art can be done.
NO GAME DESIGN?
"I want to do the environment concept art, just so I I can make all the required assets & dress the scene...
Huh? Where should I base the concept art from...?"
So how do I do the concept art if there was no game design at all... I just had to come up with some game design, which was just an answer to a combination of "what happened here" and "What player's going to do".
Some presentable concept art
SYSTEM
I always wanted to try fez-like game feature, which cooperate of a very low focal-length (10~15) perspective camera, to give a bit 2.5-sh style. In this gif I try doing fez like mechanic and apply Cinemachine camera. In the end this feature we have to crap it because I couldn't get fade-in out work.
MANAGEMENT
I've to prepare or plan, all the features that'll be use in the project. Which, I overscope myself and there's non of R&D process to be made.
CINEMACINE
This project is using URP because I've wanted to try Unity's graph features which are Shader Graph and Visual Effect Graph.
I need to set up some feature to get some testing going.
-
character controller
-
mockup animations
-
basic camera controller
-
post processing
-
scene collision
SYSTEM
I always wanted to try fez-like game feature, which cooperate of a very low focal-length (10~15) perspective camera, to give a bit 2.5-sh style. In this gif I try doing fez like mechanic and apply Cinemachine camera. In the end this feature we have to crap it because I couldn't get fade-in out work.
At that time I have to designed feature that'd detect if mesh is covering our player character. So I've designed some random bullshit the cooperate between collider detecting in-out of certain area + fade shader controller ( spoiler alert, it've work to some degree, but I think my programming skill at that time is shit so I'd cause lots of bugs and stop working )
VISUAL EFFECTS
The approach I take on maing shaders is, if there's needed graphic feature, I'd make it that instant. All of the shader in this project used I've made it all of it. Which there are, fade dither, dafe dither + alpha clip, alpha clip, normal opqaure, Flipbook.
For Visual Effects, for fire and glow effect, I've used flipbook + bill-board a
For character Sprite animation, I've used quad + material that uses a flipbook shader, then map each animated frame in the Animation timeline. The reason I've used quad is to achieve a more 2.5d-ish style and to achieve realistic real-time shadow.
If possible, I'd like to try write custom shader for shadow, to avoid using real-time light as much as possible, since it isn't that much obvious
test real-time shadow ( beary see )
2D ASSETS
If possible, I'd like to try write custom shader for shadow, to avoid using real-time light as much as possible, since it isn't that much obvious
ENVIRONMENT ASSETS & ASSEMBLING
modular assets
I mentioned above that making an environment concept art is essential to address the required art assets, which is this step, the making of 3D environment assets. I approach this by making modular 3D assets and assembling everything together in the engine.
change floor tile variation and test light baking in Unity
assembled 3D assets + lighting + post-processing
My Conclusion :
I've failed to complete the whole game, but I'm glad that I've got a chance to experiment with stuff out of my comfort zone, especially the programming and game design part. As a result, I've gained perspective through another part of game development other than art.
I've also found out which part of game development I'd be interested to deep dive into, which probably is the environment art.
I think that the management part of game development is underrated, and is an actual thing that will guarantee success in a project. In this situation, I'm a solo developer, and I shouldn't prepare for every week's presentation, while also looking out for how much I needed to work to finish. Unfortunately, I've found out after going into the game development industry, that normally doing that amount of work is ridiculous.
Video showcase real-time gameplay in engine
Anyway, this is a video showcase the project, real-time rendering.
*if the video is unavailable, some media don't play, refresh the page
Index
-
System
-
Management
-
Art Assets
-
Cinemachine
-
Visual Effects