Ask HN: Best stack for building a tiny game with an 11-year-old?

[richardstahl]

I want to make a simple game together with the 11-year-old daughter of a friend during a weekend where they stay over.

I have a Mac and Claude Code Max and Codex, so I am equipped to create AI-slop. I’m happy to do some setup and pre-wiring. Mainly I want her to understand some basics and feel the joy of building something visual in a few hours. Based on historical experience it will have to be something with pink unicorns.

I tried Godot, but it felt like too much complexity for this use case. If we do a bit of pair programming then using Godot would take too long to iterate or explain concepts. I looked at https://github.com/Jibby-Games/Flappy-Race for instance but do not think I could make that work with her in an afternoon or two. I was also unsure how to get or manage game assets.

Would you recommend Godot, Scratch, PICO-8, JS in the browser (p5.js), or something else?

Especially interested in replies from people who’ve actually made games with kids around this age.

[codingdave]

I made games with my kids when they were that age.

We did it with cardboard, poster board, markers, tape, etc, and built a real-life model/mock-up of the game first. Then we let them play it and only when they actually liked it did we code it.

I found that while some kids want to learn to code, most kids don't care about the code and want to make a game. So if you jump straight to code, you actually make them lose interest. Make the coding a phase 2 once they already know the game is fun.

[TacticalCoder]

I happen to have a 11 y/o daughter too. I teach her coding not because I want to make her a dev: but to me it's important that she gets, early on, the mindset.

So "gamified coding" it is. Online platforms with simplified source code (but still actual code), making little animals move on the screen, by using code, to solve challenges. She already did more than 300 such challenges and understood loops and variables. She begins to understand how functions do work, side-effect, pure functions, etc.

But the thing is: she's 11 y/o. I'm the boss. A nice boss, a gentle listening boss. But I'm the boss. I get to decide whether she can spend 16 hours per day in front of screens playing or not (she can't). I get to decide how many time she can play out in the street with the neighbour's kids (as I type this she's doing just that) and how many hours she does maths, code, etc.

Then I apply concepts between maths on paper and code. For example she loves having one variable named "O_o" as a joke.

This shows a great understanding of what a variable is (although a poor choice in naming it!).

Then we do maths, on paper, and I literally generate her exercise on paper sheet with:

    O_o = 3 * x + 2 * y + 7

    Find me the value of O_o for the following values of x and y.

She loves that: suddenly she sees, on paper, that same O_o variable she sometimes uses in the code as a joke. And she sees me doing the joke.

She'll be a rebellious teenager soon enough that, as of now, I still get to decide ; )

P.S: at 11 y/o she doesn't have a phone yet either. That shall come soon enough too.

P.P.S: space repetition too. For maths for example... I see where she's struggling, then it's space repetition mode. I generate her custom exercises and three times a week it's hammering.

[falcor84]

> most kids don't care about the code and want to make a game

Funny how this is somewhat of an endorsement of vibe coding a prototype.

[richardstahl]

Thank you, good advice. I am not yet sure of going the full monty of doing a complete real-life mock-up, but maybe rather focus on a conceptual understanding of how things should wire together.

As a bonus I also think that if we create conceptual game assets in the pen and paper way first - say a 60 second drawing of her beloved unicorn - she can use some quiet time later to make it more elaborate.

[brudgers]

[random advice from the internet]

Focus on fun and play.

Go where that leads.

Be a peer. Don’t lecture.

[cableshaft]

Love2D is pretty good, and what I'm using to develop the game I'm working on (Claude works pretty well with it too, I've been assigning it some tasks on Github).

Pico-8 is a bit simpler and is also based on Lua.

There's also the Playdate console. Anyone can make games for it, and it's also based on Lua (and has its own low-code Pulp engine you can make games with also). That would be a fun way to make games and have a console to play them on.

The console isn't cheap though, and it's easy to break (I dropped it on accident from about 2 feet up and its liquid crystal display cracked and leaked and had to be replaced, at almost the full cost of the console again), so they'd have to be careful with it. But it's fun!

[ChristopherDrum]

I'm a big fan of Pico-8. Great community ready to help beginners and pros alike. The feedback loop between its built-in tools is really satisfying, plus its system design forces us to scope our ideas down, though there is surprising depth once you've gotten used to what it can do.

[turtleyacht]

Scratch 1.x (1.4) is nice if you already have some books bought or borrowed. Very quick to get something running and everything is visual. (Avoid any games that need clone, or duplicate the functions.)

Two approaches: either build as many games as you can, or create a couple physics-based games that can be dissected and modified for insight on mechanics.

Penguin Jumper and Tower Smash highlight Scratch's ability to simulate game physics. They are from Code Your Own Games! Twenty Games to Create with Scratch (2016) by Max Wainewright.

Likely not much can be done on a weekend versus months; regardless of the technology, have at least one "Wow, amazing" thing from community or Steam to show what's possible.

[simquat]

I would recommend Scratch, the blocks prevent syntax errors, plus there's a simple editor for drawing sprites and backgrounds — my niece used to play with it paired with the book "Scratch programming playground" if I recall correct —.

I'm working on a visual app builder, Breadboard[0], we're not targeting education neither games, but here an example of a tic tac toe game made with it.

Workspace: https://app.breadboards.io/shared/c3c4dc7b-78f7-4e0f-b89f-cc...

Exported app: https://black-creek-7582.breadboards.app/home

[0] https://breadboards.io

[giantg2]

I would make a cli game. Something like generating a random number between 1-100 and having the user guess. The game tells you higher, lower, or correct. Then let her try a few times. Then you tell her you can do it in 7 guess every time. Explain the math about alwasy closing the midpoint of the remaining range to cut the field in half. It's like a magic trick based on math that teaches basic syntax.

[d--b]

I think this is right.

I was going to suggest the same game, guess the number. But instead of doing it in CLI, do it on her class calculator. If she's 11, she may already have something like a TI 82. These come with some form of BASIC (at least it had 30 years ago).

I remember that I did that about 30 years ago, and then because I had done it, I spent hours playing it.

Then if she likes that, you can make a 2 player tic-tac-toe, or a play-against-the-computer paper-rock-scissors. The fun is to start simple and then modify the game slowly. For example, in the guess a number, you could have funny answers for special numbers.

or a tame-the-unicorn game that's a bit like a tamagotchi, you can feed her, play with her, let her sleep, that kind of stuff, with very simple counters. In the end you can to ride on her back and be her best friend forever.

Anything that's turn-based, where you have very simple inputs that go in variables, check the variable against the game state, and then change the state in a basic fashion, and output something as text, and repeat.

Anything with a graphical game loop is going to be too much imo.

[al_borland]

If she hasn’t done anything like this before, and you’re hoping she learns something and has something to show in a few hours or over a weekend, I would use Scratch and avoid AI.

Scratch lets the syntax get out of the way, so she can focus on the concepts of what the various puzzle pieces do (loops, conditionals, variables, etc). Feedback is immediate, which makes even simple actions quickly rewarding.

If you want to do some pre-wiring, you can make some stuff in there to get a feel for it and provide some examples.

I made a game in Scratch for my nephews a few years ago. I models it after a web game that made the rounds back when I was in high school that I found fun. They liked it, and the oldest one almost instantly jumped into the code to start tweaking variables to make the game more chaotic. I think he had a little exposure to Scratch at school. He was probably around 11 or 12 at the time.

[cupcake-unicorn]

When I was a kid I made a bunch of games myself using Klik n' Play. The modern successor to this is Clickteam Fusion and there is another nocode game engine called Gdevelop that may be similar.

I liked it because it focused on events and had a lot of stuff that was built in like movement I could assign to a sprite etc. After a while I started running up against the limitations of it because it's an event based system and not a programming language. But for that time that was what I really wanted to do, I wanted to make a bunch of small fun games and I was able to do that without worrying about coding, and then I was able to learn that later at my own pace while still having all the fun from the small games I made in the meantime.

It may be different these days because AI can help fill in the gap but that was my experience

[nananana9]

I started with GameMaker when I was 10 and to this day it's the most beginner-friendly environment I've used. These are the programs kids have actually used to make games.

Your good choices are GameMaker, Construct or an old version of Adobe Flash. Those are the tools kids have actually used to make games. RPG Maker is a honorary mention. The only modern one that I'd consider is Roblox, but that has community baggage, and is a bit more difficult than them.

Unfortunately even the "easy" engines (Godot) nowadays assume you more or less programming, the scene graph, components... The old ones are unmatched.

[jay_kyburz]

For an 11year old, Stick with scratch if you ask me.

Take syntax out of the equation and learn basic variables and looping.

Just have fun moving some characters around the screen and have some numbers go up!

Keep it really simple. There could be be a lot of new concepts to learn.

[lacker]

I've been making games with JS in the browser with my kids, ages 7-13. Very simple games, the sort where we can just use emojis instead of real game assets. Even just building a game inside a Claude Artifact is pretty fun.

The nice thing about JS is that there is not very much overhead in setting things up, debugging weird things, restarting.

[benoau]

I would go JS/HTML5 and just do something really basic like memory card game so the programming will be very simple for you and the path to a playable prototype is very short, maybe not even 100 lines of code, nothing to set up, nothing to google. Once it's playable you can focus on polishing how it looks which will probably be more fun.

[bilsbie]

Would you use a js game engine or just vanilla js?

[benoau]

Just vanilla JS unless you've got prior experience because any engine you use is going to have a setup process and bootstrapping code and a learning curve for you that will eat into your time. Across the weekend you might only really have a few hours to dedicate to this project and to hold their attention.

Using the "memory" game as an example, do you want the problems you solve to be how to shuffle the cards in a random order, or do you want to be solving why the cards all positioned weirdly because PhaserJS defines an anchor "origin" point in objects and by default that's x 0.5 / y 0.5 which means 50% width / 50% height aka the center of the object so you need to either set their origin to x 0 / y 0 or factor that into their position by subtracting half their width and height, and their width and height has scaled and unscaled values too width vs displayWidth... and of course if you're using a group for the card's display objects, that class does not support setting the origin.

[textwizhub]

There are several engines what doesn't require coding but more like a drag and drop functions. I am not sure if you are familiar with Construct and GameMaker, but you can check them and see if they works best in your case with your daughter.

[morbicer]

Little.js

Small yet covering all basic needs including sound, performant, great examples that you can use as starters (breakout, platformer, puzzle)

https://github.com/KilledByAPixel/LittleJS

[deterministic]

Just make a simple CLI adventure game:

1. Print "You are in a dark dungeon. There is a door ..." 2. Print "Your options are: 1. Open the door 2. Exit 3. Turn on ..." 3. Read a number as input 4. Goto (yes goto) similar code depending on choice.

[sminchev]

You have Claude Code. Use it ;) It will be really a good exercise for her to train how she express herself. To cleanly and correctly explain what she wants. And you can guide her, give her some tip and tricks. Once you have something really simple, you can try to show her some peaces of the code, just to show her how thinks looks like, without going in to details how it works and what it does. If she at some point get curious, she might come and ask herself, and this is where the magic and love of code will happen ;)

I would propose a simple program with a few steps, it can be a console one, to draw a circle, to draw a rectangle, to draw a pink unicorn ;)

Happy, for me, I don't have such problems. My girl wants to be football player.

[uyzstvqs]

Love2d, or maybe play around with Garry's Mod.

[xeonmc]

Love2d is the best

[wvbdmp]

Playstation Dreams

S&box

Roblox

No?

[sph]

For the love of god let her learn like a human child without the crutch of AI hallucination machines.

[asxndu]

Definitely Game Maker.