The AI Challenge: Coding Tetris Without Human Help

The AI Challenge: Coding Tetris Without Human Help

Table of Contents

1. Introduction
2. Setting Up Custom Instructions
3. Installing Unity and Getting Started
4. Creating the Sprites for Tetris
5. Spawning Tetrominos and Controlling Blocks
6. Fixing Issues with Border Detection and Rotation
7. Taking Control and Making Decisions
8. Adjusting Prefabs and Aligning with the GRID
9. Implementing Collision Detection and Wall Kicks
10. Troubleshooting the Spawning Script
11. Trusting Intuition and Finding the Solution
12. Conclusion

🎮 Can AI Code Tetris Without Human Help? Let's Find Out!

In this video, I wanted to test the capabilities of CH GPT, an advanced AI language model, by challenging it to code the legendary Game Tetris. The catch was that I wouldn't write a single line of code myself; it would all be done by the AI. Little did I know that this experiment would take me on a rollercoaster ride of triumphs and challenges, teaching me valuable lessons about the limits of AI and the importance of human intuition in coding.

1. Introduction

Coding a game like Tetris requires a deep understanding of game development principles and various components, including spawning, collision detection, rotation, and movement. I gave the AI a simple task: code Tetris from scratch. Instantly, CH GPT provided me with a detailed 10-step plan and the initial blocks of code required.

2. Setting Up Custom Instructions

To ensure smooth communication with the AI, I set up custom instructions that allowed me to specify my requirements and expectations clearly. I instructed the AI to slow down, explain each step thoroughly, and avoid rushing through the project. Controlling the Prompt and managing expectations were crucial in defining the direction of the project.

3. Installing Unity and Getting Started

Before diving into the game development process, I needed to install Unity, a powerful game development platform. It had been six years since I last coded a video game from scratch, so the installation process brought a mix of excitement and anticipation. I also decided to skip the Unity tutorials and jump directly into the project.

4. Creating the Sprites for Tetris

Sprites are essential visual elements in a game like Tetris. I needed sprites for the seven tetromino shapes, the border, and the background grid. After sourcing textures from the web, I polished them in Photoshop and imported them into Unity. However, I made a rookie mistake by not sticking to a consistent graphic style. This caused visual chaos and misalignment, requiring hours of Photoshop work to rectify.

5. Spawning Tetrominos and Controlling Blocks

To make the tetrominos spawn and enable player control, I relied on the AI to generate C scripts. The AI provided the necessary scripts, allowing the blocks to spawn and move within the game grid. However, there were initial issues with spawn positions not aligning with the grid and blocks occasionally clipping into the border. The rotation mechanic was also flawed.

6. Fixing Issues with Border Detection and Rotation

Underneath the surface, a stew of problems started to brew. The spawn positions were misaligned, the border detection was faulty, and the rotation mechanic was completely messed up. These issues traced back to my overreliance on the AI's instructions and failure to take charge of the decision-making process. It became clear that I needed to thoroughly understand every detail and make informed choices.

7. Taking Control and Making Decisions

Realizing the importance of my own role in the project, I decided to take complete control and start with fresh empty scripts. It was vital to balance the contributions of the AI with my own knowledge and expertise. I updated the custom instructions, giving the AI clear guidelines to follow and preventing it from attempting to do everything in a single prompt. I replaced the old prefabs and recreated them using individual 2D squares for precise control.

8. Adjusting Prefabs and Aligning with the Grid

To address alignment issues, I manually adjusted the prefabs so that each tetromino perfectly aligned with the game grid. Getting rid of the old border and grid sprites, I created a new playing field using Unity's tiled draw mode to ensure perfect alignment. These changes brought a sense of consistency and visual harmony to the game.

9. Implementing Collision Detection and Wall Kicks

Collision detection is a crucial aspect of Tetris gameplay. With the AI's script for collision detection, the movement was halted just before the border. Although an improvement, this approach still needed refinement. By reducing the size of the overlap box, I eliminated clipping issues. Furthermore, I incorporated Wall Kicks, a mechanic that prevented pieces from crossing the border by kicking them in the opposite direction.

10. Troubleshooting the Spawning Script

As the project progressed, the AI struggled with the spawning script. Although the script was relatively easy to write, making it work properly posed a significant challenge. The AI instructed me to create a public variable and drag the spawner game object into it, but Unity rejected this approach. After hours of frustration, the AI finally provided a working solution: Unity does not allow referencing game objects directly in prefab assets.

11. Trusting Intuition and Finding the Solution

Faced with the dilemma of new tetrominos not spawning, I decided to trust my own intuition and analyze the code. It became evident that the AI's method for spawning conflicted with the collision detection, causing the new tetrominos to never actually collide with the bottom border. By making crucial changes to the code structure, I finally achieved success, and new tetrominos began to spawn.

12. Conclusion

The journey of coding Tetris with the help of AI was filled with ups and downs, insights, and frustrations. While the AI proved to be a valuable tool, it highlighted the importance of human decision-making and the need to balance the capabilities of AI with our own expertise. CH GPT showed promise and potential but also underlined the challenges and limitations of AI in complex coding tasks.

👉 Join my community to access all the code, prefabs, sprites, and software developed in this video. Don't miss the opportunity to be part of the AI revolution and Shape the future of technology.

Highlights:

• Challenging CH GPT to code Tetris from scratch
• Setting up custom instructions for clearer communication
• Installing Unity and diving right into the project
• Creating visually appealing sprites for tetrominos
• Troubleshooting issues with spawning and block control
• Addressing problems with border detection and rotation
• Realizing the need for human decision-making and taking control
• Adjusting prefabs and aligning them with the game grid
• Implementing collision detection and wall kicks
• Troubleshooting the spawning script and finding a solution

FAQ

Q: Can AI code Tetris without any human help? A: While AI can assist in coding tasks, human intervention and decision-making are crucial for successful completion, as demonstrated in this project.

Q: What were the main challenges faced during the coding process? A: Challenges included aligning sprites, fixing collision detection and rotation issues, and troubleshooting difficulties with the spawning script.

Q: What lessons were learned from this project? A: It became evident that relying solely on AI instructions can lead to unexpected issues. Balancing AI capabilities with human knowledge and intuition is essential for successful coding projects.

Q: Where can I access the code, prefabs, and sprites used in this project? A: All resources, including code, prefabs, and sprites, will be available in the creator's community. Please refer to the link in the description for more information.