@LudensLudonauta Sounds interesting but are they all for 2D games ?
English
Ben H
6.5K posts
I think I finally managed to come up with a great game design service to offer for indies.
I call this "Gameplay Map", a collection of mechanics that build up a game.
Each entry classifies the feature in my taxonomy framework and gives insights on when to use it and synergies.
English
@slimbuck7 @misaki00258 @willeastcott Hi @slimbuck7 I would love some advices on this - had some issues too with splat transform , I’ve dmed you
English
@misaki00258 @willeastcott Perhaps I can help @misaki00258 ?
splat-transform needs flags based on the scene to generate good collision.
English
Ben H retweetledi
Blackrock CEO Larry Fink said:
"The United States 🇺🇸 is short power, short compute, short chips. There are going to be shortages in all three. We just don't have enough compute power right now."
WOLF@WOLF_Financial
LARRY FINK JUST PREDICTED A NEW ASSET CLASS The BlackRock $BLK CEO laid out where he sees the bottleneck of the AI buildout at the Milken Institute. His framing of the problem: "The United States is short power, short compute, short chips. There are going to be shortages in all three. We just don't have enough compute power right now." His prediction for what comes next: "I actually believe a new asset class will be buying futures of compute." Compute futures do not exist as a standardized tradable contract today. Fink is calling out the emergence of one.
English
@willahmed @WHOOP Hey @willahmed i'm a long term whoop user and I love the product, but maybe instead of whining about people copying you, you should actually build new - better features, and get better hardware, you're late.
Also:stop gate-keeping features behind higher tiers, that's ridiculous.
English
@bilawalsidhu I'm thinking the method used by Hyworld 2.0 (see image), might be a good solution. They only released stage 4 for now so i'm waiting to test it..
English
@amentialinked You have world labs and spAItail as the obvious candidates - but you will be semi manually expanding or stitching scenes together to get something of this scale
English
Big indoor scan - fully explorable in real-time on a browser.
Distribution of 3DGS is pretty much solved.
Blocker is still large scale capture - you need an expensive LiDAR + RGB scanner to get results like this.
360 video is still hard to pose indoors.
English
Ben H retweetledi
Jack Clark Co Founder of Anthropic:
“I'm writing this post because when I look at all the publicly available information I reluctantly come to the view that there's a likely chance (60%+) that no-human-involved AI R&D - an Al system powerful enough that it could plausibly autonomously build its own successor - happens by the end of 2028.
This is a big deal.
I don't know how to wrap my head around it.”
Jack Clark@jackclarkSF
I've spent the past few weeks reading 100s of public data sources about AI development. I now believe that recursive self-improvement has a 60% chance of happening by the end of 2028. In other words, AI systems might soon be capable of building themselves.
English
The synthesis of immersive 3D scenes from text is rapidly maturing, driven by novel video generative models and feed-forward 3D reconstruction, with vast potential in AR/VR and world modeling.
While panoramic images have proven effective for scene initialization, existing approaches suffer from a trade-off between visual fidelity and explorability: autoregressive expansion suffers from context drift, while panoramic video generation is limited to low resolution.
Stepper, a unified framework for text-driven immersive 3D scene synthesis that circumvents these limitations via stepwise panoramic scene expansion.
Paper Title: Stepper: Stepwise Immersive Scene Generation with Multiview Panoramas
Project: fwmb.github.io/stepper/
Link: arxiv.org/abs/2603.28980
English
Ben H retweetledi
An epic Paris of the future, illustrated by Moebius, aka Jean Giraud.
English
LDL-lowering drugs & Alzheimer’s risk — what the data actually shows:
• Statins: most evidence, largely neutral-to-protective (meta-analysis of 21 studies, 1.2M people: ~32% lower AD risk)
• Ezetimibe: one striking observational signal, but Mendelian randomization is null
• Bempedoic acid: no direct data yet, appears safe
• PCSK9 inhibitors: one MR study flagged possible ↑ risk, but RCTs (EBBINGHAUS, ODYSSEY) showed no cognitive signal
Bottom line: lower ApoB, lower CVD risk.
The dementia differential between agents is small and uncertain — makes sense to pick based on tolerability and how much ApoB lowering is needed.
English
@haval_zebari @theworldlabs @LensStudioDev Is there any good way of automatically building collision and navmesh for Gaussian splats ? Any recs ?
English
Testing Gaussian Splatting from @theworldlabs (Marble) in @LensStudioDev and the spatial depth is unreal. 🤯 Bringing hyper-realistic worlds to mobile AR is the future, but it’s not without its technical "hacks." 🧑💻
(1/3)
English
開発者(quatercore氏)によるUE5を用いた自動都市生成の進捗を公開しました。
BP_ProceduralMapによるプロシージャルな街路生成に加え、街区の分割、路地裏や中庭のブロッキングまで自動化している。
都市の密度や構造を動的に制御する技術的アプローチは、大規模な環境構築の効率化に活用できそう。
日本語
@johnathanchewy I wish but can’t without the blue tick , can you dm me ?
English
@amentialinked sure, would love to! feel free to send me a dm :)
English
1/ I’ve been building a simulation SDK where an orchestrator agent writes the world.
Personas driven by individual LLMs step in to live, act, and interact inside it.
Sharing more details down below!
English
Since we don't use static baked meshes, the scenes are easily editable. We've already given access to a few early robotics partners. Reach out to integrate it in your stack: gizmo.antimlabs.com
English
We're launching early access to Gizmo, our automated sim creation tool. From text and/or image inputs, our agent generates SimReady assets and scenes from dimensioned primitives, with correct affordances and articulation.
English
English
@SpAItial_AI AI has finally opened up Echo-2, their world model, for testing but how does it compare to Marble from @theworldlabs ?
I made a quick comparison using the same input image and at first glance, the worlds generated by Echo-2 feel more real.
I am not sure if it is about the scale and proportions or the way the scene looks, but Marble has a more cartoony feel.
Surfaces are certainly more polished in Marble on the other hand.
I also noticed that while Marble tends to open doors everywhere and create more connecting points, Echo-2 geometry is less symmetrical and more closed (not in a bad way).
I would LOVE to learn what type of data has been used for the training of these models 🤔
English
@JosvdWest @worldlabs I'll be honest, almost 1 $ for each generation (or more) seems to be a bit overpriced. Especially since you can run Matrix3D, or even soon Hunyuan Hyworld 2.0 for much less on inference..
English
Picking up Devex at @worldlabs. If you've shipped something on the Marble API — or you've tried and given up — DM me. I'll be in /feedback for the next two weeks reading every transcript I get. Working on making our APIs easier and better!
World Labs@theworldlabs
We're excited to be rolling out two model updates today! Marble 1.1: Improves lighting and contrast, with a major reduction in visual artifacts. Marble 1.1-Plus: Our new model built for scale. Create larger, more complex environments than ever before.
English
Learn more about the expand feature in the tutorial here:
youtube.com/watch?v=SmUlqM…
YouTube
English
Expand is now available to everyone!
Extend your world in any direction you choose: around corners, into rooms, and beyond what you can see 👀
English
Ben H retweetledi
This is so weird, why is Google, the world's biggest search engine, using another startup's search engine?
Exa@ExaAILabs
We're excited to partner with Google to offer Grounding With Exa inside of Gemini models! Using Exa's agent-first search, Gemini models can now access billions of websites, technical docs, papers, people, companies, and more. 10^18🤝10^100
English