Post

"LLMs can help non-programmers with no technical background get a foothold in doing something that used to require a LOT of technical knowledge to get started in, putting them in a better position to learn the technical stuff over time by providing a less steep learning curve" feels uncontroversial?

I see from the rest of the thread that the context is video games, and my answer is that there were already much better options for getting started in game design with no background in programming before the LLM boom. There's GameMaker for one, and genre-specific tools like RPG Maker and Inform 7.

Regardless, vibe-coding is a bad way to get into game design. The way to start is still to pick a language/platform and learn it enough to make a nice-feeling version of Pong from scratch. Getting your own Pong to feel right is the best first step to getting more complicated games to feel right.

You cannot seriously believe you will learn more about game design by cloning pong than by taking your own idea for a game, seeing it realized, testing it, and iterating on the design.

Not what I said. I said this is a better place to START if you're at square one, because it's a low-barrier to entry way to learn what goes into a game and how to work with the pieces. The reason people think LLMs are good for beginners is because they think they need to dive into the deep end.

I'm kind of surprised if you haven't heard the advice "Start with Pong," "Start with Breakout" etc. before. Obviously you learn more in Game Design 201 than Game Design 101, but we're talking about newcomers who don't already know where to begin, right?

No, I've heard the advice, it's just. Y'know. Shitty advice? If you want to learn the fundamentals of game PROGRAMMING, sure, go implement Pong. But if you want to build up game DESIGN skills, that won't really do a lot to help!

If you ever meet a game designer and talk to them and ask them how they got started making games, do you think they'll tell you "well first I cloned Pong and that was when I knew I wanted to be a game designer"? No!

I first edited the code of an Asteroids game to add a shield meter and button, the absolute first thing I remember ever doing with code. Then I made my own (god awful) Asteroids game from scratch. That was when I knew I wanted to make games. How many game designers have YOU talked to?

Just recently a friend of mine made his first game. It was a breakout clone. It was pretty terrible, and he loved making it himself and is extremely proud of it. I bet he'll make more games in the future, and have a little bit better appreciation of the games he plays from now on.

I would also point out that the experience you had was not CLONING asteroids, it was experimenting with altering the rules. The point I keep trying to make is that LLMs help people get to a place where they are experomenting with the design of a game they care about faster.

Did you not read the rest of the bskeet? Or did you stop halfway through?

No, I was making a point about a part of your post, because you were describing the first thing you did. You said you then went on to build a full Asteroids clone. This is very cool, but I don't think it changes the point I was making.