#kalaha

2025-04-14

TL;DR: I used Google Gemini 2.5 Pro to create what might be the strongest (and perhaps only) app for the Namibian board game Hus. Gemini wrote ~99% of the code, cutting development time from weeks to hours, despite minor issues with specific rules and refactoring. My focus shifted from code quality to basic checks. Early analysis suggests player 2 has a winning advantage in the standard Hus rules on a 8x4 board.

With the (massive) help of Google Gemini 2.5 pro (experimental) I present: the probably strongest Hus playing app on this planet. Hus is a board game stemming from Namibia.

(okay, seems to be the only one, according to googling, and Gemini took a single prompt to implement a basic Alpha-Beta tree search, that works very well for a game where players have something like 10 moves, and complete information).

agsteiner.neocities.org/Kalaha

Experience from working with Gemini on this:

- it could code 99%, making virtually no bugs. I changed only about 10 lines of code of >1000 lines of Javascript
- we did have a longish dispute because it wouldn't want to understand how a game where both players move their stones counterclockwise works.
- Creating a second flavor of the game that works a bit differently didn't work so well, so refactoring is still not its strength.
- Overall creating four independent little apps (one playing with one ruleset, the other with a slightly different ruleset, and two simulations to do fast engine-vs-engine match to get strong game logs) took me 2-3 hours, instead of 2-3 weeks that hand coding would have taken.
- I notice that I am moving away from trying to understand the code, and wanting beautiful code to something like "developer due dilligence" (did a hacker compromise the AI and code a virus into my app), and not caring too much about how the code looks.

Also from a lengthy 12-ply self-play it seems that this variation of the game is won for player 2. First evaluation is "player 2 wins 13 stones", which is pretty decisive already.

The original variation where you don't capture stones on the row behind the inner one gives more balanced games.

#hus #kalaha #mancala #googlegemini #gemini25pro #llm #ai #development

Screenshot of the app that the AI wrote based on my specification. I'm losing, again.
2025-04-11

Spent two hours with Gemini 2.5 creating that African bean game, complete with an AI that can beat me.

Challenges: the AI faced issues with the fact that both players play in their own area, each counterclockwise.

Actually this was the only manual change I needed to do - rename the upper row cells from 7..0 to 0..7. Couldn't bother to get the AI to do this. But apart from that it was just a (3 times refined) prompt, and one more prompt to implement a simple AI (find the move that maximizes AI stones - player stones), no alphabeta or anything.

Want to try it out?

agsteiner.de/Kalaha

(need to add the rules someday).

If I would have had to code this myself, it would probably be a 2 week project. For these small apps, the current top notch AIs give you orders of magnitude of speed. Of course this massively declines (still) with the size of the project.

#gemini25pro #kalaha #mancala #aicoding #development

screenshot of the game Kalaha. All the nice colored stones are in the AI's part of the game.
Spectrum Game of the Year 1988SpeccyWorldCup@oldbytes.space
2024-11-20

ROUND 1 Poll 195

Magic Worm (Lyndenhurst)
Shadowfax (Postern)
Space Crystal (Quark Data)
Kalaha (EMM Software)

1983 GAME OF THE YEAR
Top two games to go through to next round

#GameoftheYear1983 #ZXSpectrum #SinclairZXSpectrum #RetroGaming

#MagicWorm #Shadowfax #SpaceCrystal #Kalaha

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst