
@Playcanvas @Snap. SuperSplat Editor, Viewer, Runtime, Tools.
London, UK
Joined October 2010
- Tweets1.5K
- Following152
- Followers1.4K
- Likes2.2K
Donovan Hutchence retweeted
SuperSplat, the home for 3D Gaussian splatting, just leveled up! 🎉
👥 Follow your favorite creators
🔔 Notifications for comments, likes and followers
🧭 A redesigned home page and search
🥽 Grab scenes with both hands in VR and AR
Here's what's new 🧵👇
Donovan Hutchence retweeted
All the spices splats I recently did can now be downloaded and used for commercial and non-commercial projects at superspl.at/user/danylyon.
This illustrates why we're adding support for alpha > 1 to SOG v3.
Under repeated rounds of decimation, alpha compensation prevents gaps from forming.
For this test, the background is set to 🔴 BRIGHT RED 🔴 to highlight gaps between gaussians without compensation.
I'm working on the @playcanvas SOG v3 proposal.
So far we have:
- Tighter data packing
- Up to 4 bands of SH
- Alpha > 1 support (compensation post-decimation)
- Per-gaussian custom/blind data
- meta.json extras field, à la glTF
Would love your feedback!👇
github.com/playcanvas/splat-…
Donovan Hutchence retweeted
Insane macro-splat of a Japanese buff-tip moth.
This is less than 10MB thanks to the SOG format for 3DGS. Every tiny detail/hair is visible. 🔎
Rendered by @playcanvas using the full power of WebGPU. ⚡️
[1 / 2]
Donovan Hutchence retweeted
One of my favourite splats on SuperSplat!
It has 10M Gaussians and is just jaw-dropping when you walk inside.
With Blender 5.3 around the corner, I'm eager to see what artists can create with amazing splats like this one, which btw is available to download on SuperSplat (link below)!
I'm working on the @playcanvas SOG v3 proposal.
So far we have:
- Tighter data packing
- Up to 4 bands of SH
- Alpha > 1 support (compensation post-decimation)
- Per-gaussian custom/blind data
- meta.json extras field, à la glTF
Would love your feedback!👇
github.com/playcanvas/splat-…
Donovan Hutchence retweeted
Generated by the insanely powerful Echo-3 world model. Rendered by the open source PlayCanvas Engine. 🤝
Donovan Hutchence retweeted
The open source splat-transform library + CLI tool from @playcanvas is growing fast on NPM. If you need to do any processing of 3D Gaussian splat data, give it a look!
Donovan Hutchence retweeted
Migrating from Three.js to PlayCanvas seems to have resulted not only in graphical improvements but also higher FPS. This is in line with benchmarks I've seen too. Wish I knew earlier.
Donovan Hutchence retweeted
🇭🇺 Budapest's Parliament Building as a 3D Gaussian splat
Crazy detail from 863K splats - but just 14.6 MB! Running in a browser tab on SuperSplat.
[1/2]
Donovan Hutchence retweeted
Snapchat now supports PlayCanvas games! 🎮👻
Introducing @Snapchat Web Lenses. Build a game with web tech, zip it up and publish it straight into Snapchat as a Lens.
Here's what you need to know 🧵
Donovan Hutchence retweeted
Powered by PlayCanvas' Gaussian splat tech. 🔥
Voxelo 3.0 also lets you publish directly to SuperSplat!
Donovan Hutchence retweeted
The @playcanvas Editor 2.35 is out ✨
🏠 A brand-new projects home at playcanvas.com/editor
📦 Everything from your playcanvas.com projects page, now in the Editor: search, stars, forks, transfers, dev logs, teams and organizations
🤖 Connect MCP from the home and let Claude Code manage your projects
Donovan Hutchence retweeted
Demos + viewers powered by PlayCanvas! 🔥
Today we are releasing JST-2, the world's first full-stack physics-native world foundation model.
Many world models learn in pixel/video space. We went back to first principles and built the entire model stack in 3D.
Generate a 3D scene in seconds and expand it in real time. Create characters with motion built in, all through one general architecture.
Donovan Hutchence retweeted
A Gaussian splat of a saffron flower. You can see it's three threads which are hand picked for the spice. Trained in @lichtfeldstudio #3dgs
Donovan Hutchence retweeted
ペトラの「エル・ハズネとファサード通り」をSuperSplatで公開しました!
旅行気分でぜひ楽しんでください~
私自身も先月の調査の記憶がよみがえってきたので、結構遺跡の雰囲気が伝わるのではと思います😉
#LichtFeldStudio #3DGS
⬇️URLは以下
Donovan Hutchence retweeted
This worked better than I expected! Replaced rasterised light probes with ray tracing, and it was fast enough to run realtime. Running in WebGPU + @playcanvas
Donovan Hutchence retweeted
WEBブラウザでタンポポの綿毛の3Dガウシアンスプラッティングをご覧いただけます。
View our 3DGS dandelion seed head in your browser.
superspl.at/scene/6ea4374d
#3DGS


















