Developing a Mahjong Game Has Been Tough

Posted on May 23, 2026

Hello, this is Yuu!

Ever since I suddenly decided that I wanted to make a Mahjong game, I’ve still been continuously working on it.

Right now, I’ve finished implementing calls such as Ron and Riichi, and I’m currently working on hand evaluation and score calculation.

As for the progress since my last post, the complexity of the game has made me play Mahjong constantly just for reference and testing.

I’m still not fully confident about all the hands and rules yet, but actually implementing them has helped me learn many small details that I probably would never have understood otherwise. In that sense, the whole process has been pretty refreshing.

Last time, even the reference implementation itself was mostly trial and error, but this time I have the very clear goal of “making a Mahjong game,” so I feel like the project has been moving forward without losing direction, for better or worse.

Recently, because of Mahjong’s many detailed rules and my own incomplete understanding of them, I’ve been running into situations where bugs I already fixed end up reappearing later, or changes in one place unexpectedly affect something else.

The codebase has also grown quite a bit by now, so even small changes can have a much wider impact than expected, which makes it difficult to relax.

There is still a lot left to do, especially regarding localization and polishing, but hopefully I’ll be able to release it someday in the near future.