Point and clicks have quite a few similar examples, the genre lends itself well to dioramas because even when it's 3D it's often a fixed camera path that you can optimize for.
Papetura also got a fair bit of attention for a while (for good reason, just look at it): https://youtu.be/ZVhtuKleLuI
Personally I love the hand-crafted "real" look these bring. It tends to consume a ton of disk space, but it looks good pretty much forever.
aeontech 13 hours ago [-]
Lumino City was the first one that came to mind for me as well!
The next one was the Fantasian, by the creators of Final Fantasy (https://youtu.be/ePFgyBtvqQU) - apparently this is finally coming to non-Apple platforms this year, at that.
butlike 3 minutes ago [-]
That Dungimon system in Fantasian seems like it may be the best new idea for JRPGs I've seen in like...forever.
rkachowski 7 hours ago [-]
Same, I was blown away watching the making of video (https://www.youtube.com/watch?v=NLv2uHxygc0) and discovering that these environments were created physically. The whole world had that soft handcrafted feel, and it turned out to stem from actual crafts!
esperent 11 hours ago [-]
Same, it's such an amazing game. And the mobile port is great too.
jtxt 2 hours ago [-]
Reminds of an early tank simulator where the driver in the cockpit saw a video feed from a camera on a gantry they were slowly driving around a diarama. https://m.youtube.com/watch?v=AcQifPHcMLE
"Real engine" for just one player at a time though.
Wow that's a blast from the past! 12 year old me was absolutely addicted to this game.
krypton2k 13 hours ago [-]
There was a nice game in this style called The Neverhood.
jermaustin1 4 hours ago [-]
This is the game that comes to my mind. I had the demo from a Windows 95 game disk. The demo was great. Never played the full version though.
croo 9 hours ago [-]
That game was made out of 3.5 tonns of clay.
9 hours ago [-]
22c 11 hours ago [-]
The soundtrack for The Neverhood was also very nice.
a3w 5 hours ago [-]
Lowdee huh
fiala__ 1 hours ago [-]
this is neat! Diorama-based games are really underrated IMO. I'm currently attempting to do something related, building a game on top of Google Street View, with 3D audio and objects embedded in the panorama.
qw 9 hours ago [-]
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
rob74 5 hours ago [-]
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"...
Charon77 12 hours ago [-]
You could do gaussian splat on the diorama picture and you can have pretty good dynamic camera movements where the player could walk around
esperent 11 hours ago [-]
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.
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
aidenn0 1 hours ago [-]
Vsync locked at 60fps on my desktop 1050Ti, which was released 8 years ago, so not too surprising.
diggan 8 hours ago [-]
In true game dev fashion, I think parent meant to do the splatting once at build time, and runtime would just use the resulting data, rather than each scene load involving a dynamic "Splatting the gausses" in order to finish loading.
13 hours ago [-]
bitwize 12 hours ago [-]
Oddly enough, some of Cliff Bleszinksi's first titles published by Epic (then Epic MegaGames) were point and click adventures like Palace of Deceit: The Dragon's Plight. They had MSPaint graphics and VB code.
This is a lovely continuation of that same tradition that takes full advantage of your analogue craftsmanship skills.
nickin190 13 hours ago [-]
[flagged]
kleiba 8 hours ago [-]
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.
Vampiero 7 hours ago [-]
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 bunch of different styles and sizes or they'll look terrible. In the simplest case you want to generate at least an entire spritesheet for a single character or object. And that's still fundamentally impossible with modern generative AIs
lpat 7 hours ago [-]
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.
jncfhnb 6 hours ago [-]
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
ksymph 3 hours ago [-]
I can't speak for how well it works but PixelLab [0] claims to support animations, rotation and other things that seem impossible.
> 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.
As an example, a nice looking one from a decade ago: https://store.steampowered.com/app/205020/Lumino_City/
Papetura also got a fair bit of attention for a while (for good reason, just look at it): https://youtu.be/ZVhtuKleLuI
Personally I love the hand-crafted "real" look these bring. It tends to consume a ton of disk space, but it looks good pretty much forever.
The next one was the Fantasian, by the creators of Final Fantasy (https://youtu.be/ePFgyBtvqQU) - apparently this is finally coming to non-Apple platforms this year, at that.
"Real engine" for just one player at a time though.
- 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"...
Does anyone on a smartphone want to try it?
The calculating power of current gen devices is honestly mind-blowing
This is a lovely continuation of that same tradition that takes full advantage of your analogue craftsmanship skills.
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 bunch of different styles and sizes or they'll look terrible. In the simplest case you want to generate at least an entire spritesheet for a single character or object. And that's still fundamentally impossible with modern generative AIs
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.
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
[0] https://www.pixellab.ai
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