"Wouldn't a working approach be to just create a really low resolution 3D world in the traditional "3D game world" sense to get the spatial consistency. Then this crude map with attributes is fed into frame generation to create the resulting world?"

That's Minecraft. It does work. And when you turn around, then turn back to where you were facing, you don't feel like you're in a fever dream, because the landscape hasn't completely shifted.

The tech is neat and I'm sure worth publishing. The rhetoric accompanying it is terribly overblown. Of course that's as likely to be their University PR department as the authors.

> Wouldn't a working approach be to just create a really low resolution 3D world in the traditional "3D game world" sense to get the spatial consistency. Then this crude map with attributes is fed into frame generation to create the resulting world?

If you go "ultra-low" resolution, that's basically Minecraft or Luanti (formerly Minetest). There some mods for Luanti that let one generate a less "squary" world by adding slopes, but that's still heavily polygonal.

> It wouldn't be infinite, but on the other hand no one has a need for an infinite world either.

Minecraft has an infinite world (with glitches when you go very far from origin due to floating point errors). Luanti's world is finite (around 64000 m^3 because 16 bits coordinates; players move at 2-4m/s usually), some people are working to make it infinite. In my opinion, you are right that nobody needs an infinite world; the argument for an infinite world is that a finite world can be a problem on popular servers but in my eyes servers where this is a problem don't use enough the available space (in particular the huge vertical space that can be used to create multiple worlds) or shoot themselves in the foot by providing players with ways to move fast (mainly transports); increasing the player's speed actually shrinks the world.

I think there is a conundrum from the players and gameplay perspective. On one hand, you want a lot of space, but on the other hand you want to be close to other players (for commerce, play together, etc.). It's not enough to have an infinite world, you also have to have the gameplay that goes with it.

Another thing to consider are the interaction of mobiles (NPCs, enemies) with the terrain. Because nobody likes an infinitely empty world either. I haven't played an AAA game recently, but a decade ago even with precomputed terrain, some mobiles would look silly sometimes with steep terrains.

It's basically a dream. Maybe that's what dreams are? World prediction engine running with no sensor input and memory in read only mode?

Not an expert, but I think procedurally generated terrain is generally fractal in nature, and is reproducible in that sense from a seed that is used in the generation. It is therefore recursive, as fractals are recursive.

A traditional neural network is a universal function approximator, however it is not recursive in nature, unless it is some sort of RNN. The transformer architecture, which this seems fairly similar to this one, is also not recursive in nature; although I believe, limited recursion can come about through CoT.

Therefore, I don't believe this could match the reproducibility, in an infinite sense, of a traditional procedural generator.

What you really want is a story telling engine informing the world model and the humans need to generate power.

I think we need devices that can produce gaussian splat videos to become as common as 2d cameras.

I think fever dream AI stuff like this would be nightmarishly terrifying in VR lol.

Still, would be super cool to try

There's so many researchers working in AI now that we can afford to explore all avenues. This probably isn't going to lead to a fully generative AI game engine, but I bet there'll be useful learnings along the way.

I think Ubisoft are leaders in that space. I saw algorithmic generation of New York style geometry for some behind the scenes video for The Division 2.

I wouldn’t be surprised if epic has similar.

Generating coherent objects in an infinite world is probably harder.

Because you wouldn't have to wait for a human to write the engine and write the game

Click to play [...the video]

People have been doing that with Minecraft for over a decade. In the old days, once you got far away enough, the terrain generation would go haywire. Lots of videos from that time period of people exploring the "edge of the world".

Personally, these were the kind of glitches which made games feel magical and "real" to me as a kid. Being able to analyze a system by breaking it made it seem so much more tangible, like an actual place I had an NTSC-sized porthole into.

That community already exists because the current version of these types of AI game engines are constantly running into a surreal edge condition since they don't track things consistently when they go off frame.

Sigh, agreed!

Why is that sad?

Here's my list of submissions: https://news.ycombinator.com/submitted?id=ben_w

Of those ten:

• 6 had zero comments, 1 had only my own comment • 1 had 7 • 1 had 17 • 1 had 148

I have no reason to think there's a nationality thing here, stuff just falls off the top fast and most people don't comment or upvote… same as with most comments themselves.

? You're making a generalization / judgment based on... the amount of comments left only two hours after it was posted, late at night in the US and early in the morning in Europe. It's also very much a meta-comment instead of something substantial about the subject at hand, which isn't encouraging discussion about the thing itself but actually draws the focus on it being Chinese, which most people weren't even aware of until you drew attention to it.

Really don't think that's the case. I don't care who makes things at first; I first want to see if they are interesting and then maybe dig deeper. Looks cool, upvoted and bookmarked to wait for the playable demo.

See also: timezones

lol had no idea it's chinese nor cared which country produced it. nationalism in any form is repulsive

was intrigued by the post but couldn't get anything to play

[flagged]

Yeah, no.

It's posted at midnight on Thursday (eastern time).

It's mobile unfriendly, hard to read, and has no videos. The other models had playable demos and videos, and they were posted in the middle of the day so we could think about it during work.

The hype wave for this stuff is going to require bigger splashes for each new model. New image-to-3D models garner a yawn, and it's going to be the same here soon.

These folks put a lot of thought into their branding (and CSS), but they kind of let the excitement fizzle as there's nothing to look at and evaluate. We just have to trust that they did things? It's a bunch of pictures of a car and green text.

It's far too late to open the paper.

Basically they just don't excel at marketing. 3/10.

Edit: I had no idea this was Chinese until you said it. The page doesn't mention names at top, and it didn't suck me into the paper.

I don't know if you're actually new here, or you've been reading HN for years, but your account is only 11 months old.

We talk about a lot of things here, but when we do talk about AI, we tend to prefer talk about things with code or papers.

This particular project has a paper. They're expecting to publish their code soon.

Here's the paper:

https://thematrix1999.github.io/article/the_matrix.pdf

You don't have to read it, but you may want to consider it if you want to learn something and/or contribute something meaningful to the conversation.

But then there would be no articles!

Why? It's a mostly democratic news aggregator site without much editorial overview, if you don't like it, downvote / don't upvote it, or write a client that filters topics you don't like out based on some keywords. You didn't need to open this page and comment on it if you don't like it, nobody's making you read things you don't like.

Yes, and that someone should be you. Be the change you want to be.

Check lobsters