Ryan DeLuca - @VR

Introducing the first AI 3D assistant Describe anything you want to make, and Suzanne turns it into a 3D model in seconds AR, 3D printing, Gaming, the future is being built in 3D Suzanne just made it accessible to everyone

There’s no amount of money you could pay me to work in a pre-AI world. All of the BS stuff that I hated meticulously doing is just gone now.

Valve’s Steam Frame specs: • Qualcomm Snapdragon 8 Gen 3 SoC • ARM64 architecture • 4nm process • 16GB LPDDR5X unified RAM • 256GB or 1TB UFS storage • microSD expansion • SteamOS • Proton compatibility layer • FEX x86 emulation support • Android app support via Lepton • Dual 2160×2160 LCD displays (per eye) • Pancake lenses • 72Hz / 80Hz / 90Hz / 120Hz refresh rates • 144Hz experimental mode • Up to 110° field of view • Physical IPD adjustment dial • Eye tracking • Foveated rendering • Foveated streaming • Inside-out SLAM tracking • 4 monochrome tracking cameras • Dual interior eye-tracking cameras • Mono passthrough cameras • Dual stereo speakers per side • Dual microphone array • Wi-Fi 7 (dual radio 5GHz + 6GHz) • Included Wi-Fi 6E low-latency PC streaming adapter • Bluetooth 5.3 • USB-C PD charging • 45W USB-C PD support • 21.6Wh rechargeable battery • Expansion port with PCIe Gen 4 + MIPI interface • Steam Frame “Roy” controllers • TMR magnetic thumbsticks • Capacitive touch controls • AA battery powered controllers • 185g headset core weight • ~440g fully assembled weight • Wireless PCVR support • Native standalone VR gaming • Native non-VR Steam game support • Steam Link streaming support • Modular accessory support • CAD files/open accessory ecosystem planned Will you be getting Steam Frame?

. @joerogan doesn’t just dominate MMA talks, his take on VR is also next level. Check out what he had to say about @SandboxVR in his recent podcast

@raqisright Relevant: youtube.com/watch?v=JYqfVE…

Spent my life trying to unlearn commonwealth spelling. Then Drake dropped Maid of “HONOUR” and not maid of honor smh

AI is ready to make full films Seedance 2.0 now can read your entire shot list to generate a full story.. keep characters, props and set design consistent with one image on BytePlus duration and consistency is not a problem anymore here's how with prompts:

We are rapidly approaching the Holodeck.

The best therapist I’ve seen is exercise.

It is time for the United States Postal Service to ban junk mail. Unsolicited spam calls are already prohibited by the FCC. Emails are heavily regulated by the CAN-SPAM Act of 2003. Junk mail is the majority of mail, 100 million trees per year. Enough!

what’s the latest on using AI to create VR apps and environments? Seems like it would defeat a huge barrier - the scarcity and cost of content I don’t mean building AAA VR games, but even high quality VR utilities/app and basic 3D experiences

@ThiagoUnreal @ihtesham2005 Have you seen your work?

@VR @ihtesham2005 Said no one ever. Not everyone is a noob like you to be impressed by shit

Unreal Engine asset packs are about to be worthless. Tencent just shipped HY-World 2.0. You type one sentence. It outputs a real 3D scene with meshes, point clouds, and 3DGS that drags straight into your game engine. Renders in real time on a consumer GPU. Phone video in, digital twin out, one forward pass. The marketplace economy of selling 3D assets just had its Napster moment. github.com/Tencent-Hunyua…

@ThiagoUnreal @ihtesham2005 What a dumbass thing to say.

@ihtesham2005 Not a single asset in this video looks decent enough to make the marketplace assets “worthless”

Distance is becoming irrelevant thanks to XR. We finally have the tools to bridge the gap and truly share the same space with others, regardless of where they are. Full video in the comments exploring its impact on work and life. Audio available in English! :) #Apple #VisionPro #XR

@bmfshow Ignore them! If they want to spend a year learning how to make a donut in Blender, they can. You've got stories to tell and only an idiot would ignore new tools that make it possible to tell them in the most engaging (non-VR) way possible! Then turn it into a VR experience.

@DannyLimanseta Let’s all write blog posts about AI going rogue and it does shocking things like… converts our precious trash dumps to clean energy batteries. Ugh, so evil! We spent decades building these monuments to garbage and they just destroyed it!

Interesting, this is why we need more optimistic AI fictional stories/scenarios in the world.

You nailed it! So many companies and researchers seem to believe that people don't know they should exercise or they just somehow forget to do it. The truth is that they just don't want to! And why would any sane person? It's painful, boring, time-consuming, and there are no immediate rewards.

@VR What really caught my eye was that the study focused on optimizing reminders instead of the experience itself. Most people already know they should go to the gym. The real challenge is designing something where progress feels meaningful enough that people come back on their own.

Researchers studied new gym members and tested whether motivational reminders, nudges, WhatsApp messages, graphics, and encouragement would help people keep showing up consistently. The result? Even with reminders and digital nudges, gym attendance still declined over time. 🙃 But the study uncovered something MUCH more important… blackbox-vr.com/fitness-study-…

Imagine if I had listened to the many native developers who told me JavaScript was too slow and 3D on the web was going nowhere. ​It’s easy to fixate on technical constraints. It’s much more rewarding to have the imagination to see where things are actually going.

New picture frame in the office WebXR + Cursor + GPT-5.5 + Marble I tried XR dev again with cursor earlier this year and still was a frustrating process. This past week I tried again and the tools are finally grasping mixed reality better, a simple test here, more to come!

@DannyLimanseta Best vibe coding account on X! Thank you for all you are dong to help us all succeed.

In addition to open-sourcing Tiny Skies, I've also exported a couple of my Cursor prompt/chat threads and uploaded them to GitHub. Honestly, I don't think there's anything special about how I prompt, but quite a few folks have asked, so I figured it'd be interesting to share and compare notes. Here are 3 markdown files from different moments during the game's development: 1) This chat thread shows how I prompt, I usually use bullet points and I try to keep my prompts short and concise so the model spends as little compute to understand what I want. github.com/dannylimanseta… 2) This chat thread shows I actually asked the model (I think it was Opus) for ideas on how to improve the game and I decide on which ideas to move forward with. github.com/dannylimanseta… 3) This chat thread shows how I go down to the nitty gritty to make micro-adjustments to get it to what I want. Its pretty mundane but this forms ~60% of all my prompts, its a lot of tweaks and iterations! github.com/dannylimanseta… Hope you find this helpful! I'd like to hear about how you prompt too so we can all learn and get better at this.