Context
I have recently decided to focus my efforts on the Go programming language. I have used it professionally before and it is a pleasure to go back to it. I really enjoy working with it.
In order to practice, I have created a very crude chess game. You can see the code on Github here: https://github.com/nakurai/go-chess-game.
I have tried several time to build the chess logic, and gave up every time. I am very happy to have completed it. I used the most naive data structures and algorithms in this logic, but that's how I was able to complete it.
To handle the UI, I used Ebitengine. The documentation is quite sparse (at least for now) but it is a very neat library. I would like to do more with it in the future.
As you can see in the video, the "AI" is for now very trivial. It just picks a move randomly among all the possible ones available.
Challenges
In terms of challenges, learning how to use the game engine while ramping up my skills in Go was the main challenge I would say. The logic for now is still quite straightforward, and having the unit tests helped a lot to detect edge cases.
Roadmap
There are many ways to improve this project, and I think it's a very good playground. On my list of improvements are:
- Make the AI incrementally smarter. I have a lot of ideas about this. For a starter, I would like to build a heuristic function able to assess the intrinsic "value" of a board in any specific state. That would help determine the next move. I know stockfish has one, so it should be feasible.
- Make the UI part of the game better. For now, it does not even acknowledge a checkmate/draw state!
- Make the art look better. I would like to learn how to use 2D sprites with Ebitengine and make the game as a whole better.
- Optimize the chess logic. For now it is all very naïve, and given the simple "AI" it does not show its limitations. However, there are a lot of possible optimizations. As the logic gets more complicated, the idea would be to keep an eye for emerging bottlenecks and go from there.
Takeaways
In my previous attempts, I tried to follow a tutorial on how to build an optimize version right away. I think that was a mistake, especially while learning more about a language, a library, etc.
"Make it work, make it good, make it fast" rules. Well, I made it work.
Comments
Post a Comment