Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Kung Fu Chess (github.com)
140 points by paladin314159 32 days ago | hide | past | web | favorite | 37 comments

Bobby Fischer would have loved this. He was constantly complaining that chess had become boring because everyone had "memorized the strategy books". That's why he invented Chess960[0].

[0] https://en.wikipedia.org/wiki/Chess960

Fischer invented Fischerandom. It was renamed later.

The idea is great, but seems the implementation (when playing a game) doesn't quite work out well from an UX perspective. I kept struggling to move pieces, while the AI was able to move multiple pieces in a few seconds.

Perhaps I wasn't holding it the right way, but it's something I'd want to play against friends in the future. Great stuff paladin314159!

Thanks for the feedback!

Are you playing on your phone? I never really solved the UX problem on mobile because of the necessary speed/accuracy of moves. The original game was only on PC.

Not the original poster, but I tried it from my phone. I found it sort of playable. Pieces mostly moved where I wanted them to. The ai does seem to be able to move more pieces than would be humanly possible at a time though. I started, I moved one pawn and in the same time the ai moved 5 pawns and a knight simultaneously. It would be fun against a human opponent using the same kind of device I think.

This was on my laptop, phone took too long to load.

Yeah. And when both(human and AI) move pieces simultaneously, the AI always gets it. (Even if I started first). Is this a rule?

I had the same problem. Was I trying to drag pieces the wrong way? I'm using a MacBook touchpad and not a mouse.

There's also something called Chezz, which is virtually the same thing and probably derived from the same original.


I used to play Kung Fu Chess by shizmoo games years ago. Their four player mode was intense!

Not sure what happened to it but here is the wiki:


Great to see!

I also recreated this amazing game a while ago because I could not find it any longer: https://github.com/PetterS/realtimechess

This reminds me of the kick-chess with dices I used to play when I was bored of traditional chess. You and your opponent would roll dices, whoever had higher number would kick a piece of its own with fingers to go bump over as many pieces of the opponent. Win the game when you kicked all pieces of your opponent off the table. Hit too hard and your finger would bleed. Hit too low and you might only kick over a single or sometime even none of your opponent pieces. Use always the same finger for kicking was one of the rules. After 3 or 4 games, your finger would hurt so much that you'd also want to stop, but there was another rule - who was saying 1st to stop playing was automatically giving 3 games to the opponent. So you wanted to either have at least 4 games ahead then stop or on a tight match you'd grind through your pain to outlast the opponent's pain. Good times.

Good job, my only complaint: Check should have some kind of impact. The king being able to move into check was surprising.

Having implemented and played this game a lot I don't think check makes sense in this game.

There is an advantage of moving into check quite often. (A check with a rook across the board is quite weak)

Better to just play until the king is captured. (This is also how regular check is equivalently played)

I’m not sure I understand. In normal chess it’s illegal to move your own king into check.

I meant that the game does not change in a meaningful way if moving your king into check is allowed. You will lose either way.

This is a variant of chess with different rules.

I get that; I was just asking the parent to explain why check rules differ in this format. (It’s okay though, another comment put it in a way I understood.)

As long as you can cover it before the checking piece can finish its cooldown, you should be fine.

Somebody should implement this in the physical form.

Here's an idea: Strong magnets (or just pieces of strongly ferromagnetic material) would be embedded at the bottom of the chess pieces; the board would use a strong solenoid on each square to fix the piece once it touches down. Alternatively, the bottom of each chess piece could have a hermetic seal ring, and there's a hole at the centre of each square that will create a vacuum to fix the chess piece in place.

The pieces could be (3D printed) in translucent material with an induction circuit with an LED inside. The LED will glow at various intensities to represent the cooldown.

The next obvious step in the escalation is a kriegspiel variant.


If 2 pieces collide while moving, the one that started moving first prevails.

That kind of makes sense right? First-mover advantage

It does make sense, but it caught me by surprise.

This is great! Nice one! I'm intrigued by the idea of infinite chess - I'm thinking a chess game that when a piece is taken it reappears on it's starting square (when it's next free).

Edit - something like what I suggested exists - https://en.m.wikipedia.org/wiki/Infinite_chess

You might also be interested in Shogi. Captured pieces can be redeployed onto the board with relatively few restrictions in terms of location. It's fun! https://www.chessvariants.com/shogi.html

Commercial Realtime Battle Shogi game was released this year for Nintendo Switch and Playstation 4. https://www.silverstar.co.jp/02products/rtbs/

Wow, that's awesome! I am bad enough at Shogi that I think I need to stick to turn-based for the time being though. :)

I think the moves should be near-instant, here you can see the pieces moving and dodge them. That's probably the less "chess-like" feature of the game.

I played a few rounds, pretty fun!

Very nice and very fun! Good job!

Though sometimes have a problem castling? Is there a specific way I need to click in order to do this?

This is amazing.

I enjoyed playing this

TL;DR: "Chess Without Turns"

Haha - wonderful. Have an upvote.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact