Live data from Hacker News

Building a Game with the Real Engine

novalis.org

21–30 of 41 posts

Re: Building a Game with the Real Engine

#21
post #18

It seems quite obvious that Gen AI will be a major change in the creation of game assets. If not stand-alone then certainly in the form of AIded (get it?) tooling. In particular there's huge potential for the indie market. If you're thinking about starting up a company, that's definitely a space where things are going to be hot IMO.

Every single game dev on Earth has been thinking that since the very first GANs, but the state of the art when it comes to generative AIs that are actually useful for asset generation is still pretty bad. Much like LLMs are still pretty bad at logic. It seems like we're plateauing in that regard... The thing about assets is that they have to be coherent with each other. You can't just generate them one at a time in a…

Consistency was for sure the biggest issue when I tried this approach last time. Also most of the images were a bit off, like having misaligned limb or some visual glitch. Sometimes fixing those details can be harder than drawing some crappy image from scratch.

I really wish there was a tool where I can select the theme, which objects I need and it would spit out some half decently looking art. I even don't care if it was some simplified, cookie-cutter art as long as it was low-effort and consistent. It could be a real time saver for prototyping.

Re: Building a Game with the Real Engine

#22
post #18

It seems quite obvious that Gen AI will be a major change in the creation of game assets. If not stand-alone then certainly in the form of AIded (get it?) tooling. In particular there's huge potential for the indie market. If you're thinking about starting up a company, that's definitely a space where things are going to be hot IMO.

Every single game dev on Earth has been thinking that since the very first GANs, but the state of the art when it comes to generative AIs that are actually useful for asset generation is still pretty bad. Much like LLMs are still pretty bad at logic. It seems like we're plateauing in that regard... The thing about assets is that they have to be coherent with each other. You can't just generate them one at a time in a…

I’ve been exploring this for a while. Icons are pretty solid with flux these days. Previous models struggled with this.

Sprite sheets are achievable or near achievable imo. SDXL was maybe good enough as a foundational model if you could leverage IPAdapter or a Lora to maintain character consistency. Although the former was never effectively tuned for styles I wanted and the latter requires some actual artist output still. Flux as a base model is far more consistent and has good enough controlnets to do sprite generation but probably still needs some a proper IPAdapter running to be stable enough for production.

Ideally it’s just a productivity tool for an actual artist

Re: Building a Game with the Real Engine

#23
post #7

You could do gaussian splat on the diorama picture and you can have pretty good dynamic camera movements where the player could walk around

Your game would go from playable on a smartphone to requiring an rtx 4090 and 64gb of RAM (ok, slight exaggeration). But it would certainly look amazing.

Just for reference, here is a real-time 4D Gaussian web viewer: https://antimatter15.com/splaTV/

Does anyone on a smartphone want to try it?

Re: Building a Game with the Real Engine

#25
post #16

I accidentally hovered my mouse cursor over the images, and saw a detailed description. It may have been generated by AI, but it's a nice touch that was not expected

I especially liked the descriptions to the eagle images, where the prompt was "Painting based on Goya’s Saturn Devouring His Son, but with a giant eagle eating a tiny headless monk in a brown robe. The eagle holds the monk’s in its giant claws. Crimson fluid leaks from the monk’s neck.", and the descriptions are:

- an eagle feeding blood to a child in a brown robe.

- an eagle dripping blood onto the hand of a youth in a brown robe; the child's hand and face are bloody and the hand only has three fingers but not because one was eaten.

- a child with no legs holding back the claw of an eagle

- an eagle with its beak at the face of a man; maybe it's eating him or maybe it's feeding him. The eagle has distorted, weirdly large human hands wrapped around the man's neck.

BTW, the alt text, the image filename and the element id are all variations of "Fuck Midjourney"...

Re: Building a Game with the Real Engine

#27
post #18

It seems quite obvious that Gen AI will be a major change in the creation of game assets. If not stand-alone then certainly in the form of AIded (get it?) tooling. In particular there's huge potential for the indie market. If you're thinking about starting up a company, that's definitely a space where things are going to be hot IMO.

> It seems quite obvious that Gen AI will be a major change in the creation of game assets.

Check out the authors discussion on the subject of 'turds' [0], image artefacts created by AI that totally break immersion as soon as you properly notice them.

[0] https://novalis.org/blog/2023-05-30-turds.html

Re: Building a Game with the Real Engine

#28

Earlier quoted context omitted.

Your game would go from playable on a smartphone to requiring an rtx 4090 and 64gb of RAM (ok, slight exaggeration). But it would certainly look amazing.

Just for reference, here is a real-time 4D Gaussian web viewer: https://antimatter15.com/splaTV/ Does anyone on a smartphone want to try it?

80-90fps on pixel 8 pro. Occasionally spikes to 3 digits for a few moments.

The calculating power of current gen devices is honestly mind-blowing

Re: Building a Game with the Real Engine

#29
post #18

It seems quite obvious that Gen AI will be a major change in the creation of game assets. If not stand-alone then certainly in the form of AIded (get it?) tooling. In particular there's huge potential for the indie market. If you're thinking about starting up a company, that's definitely a space where things are going to be hot IMO.

Every single game dev on Earth has been thinking that since the very first GANs, but the state of the art when it comes to generative AIs that are actually useful for asset generation is still pretty bad. Much like LLMs are still pretty bad at logic. It seems like we're plateauing in that regard... The thing about assets is that they have to be coherent with each other. You can't just generate them one at a time in a…

I can't speak for how well it works but PixelLab [0] claims to support animations, rotation and other things that seem impossible.

[0] https://www.pixellab.ai

Post reply on HN