Index
Every breakdown, Skill and Prompt. Filter by type, portal or tag, or search for a tool.
117 pages
How to keep one character recognisably the same across 2D images, a reused 3D hero model and AI video clips: a master reference sheet, a fixed identity block, the reference features of current tools, and a drift check on every output.
Turn playtest words like floaty, slippery or weak into measured numbers (acceleration and stop times, coyote time, input buffers, camera damping, hit-stop, screen shake) and tune them with an AI model one or two values at a time.
The full path from one concept image to a game-ready 3D model with Tripo, Meshy, Rodin or Hunyuan3D: input image rules, picking candidates, decimation and retopology budgets, UV and texture fixes, scale, pivot and export.
Build AI video from designed stills instead of text: start and end frames, 5–10 s clips chained into a sequence with identity and light held steady, the best takes picked, then upscaled and interpolated.
How to split game work between separate AI agents (planner, builders, playtester, reviewer), connect them with task cards, hand-off notes and screenshot evidence, stop them undoing each other, and keep token costs in check.
How to turn photos of a real place into a playable level, with a desk race track as the example: capture and scale reference, blockout, modelled, generated or scanned props, collisions and the racing line, and lighting that keeps the real look.
The release checklist between a playable prototype and a link anyone can open: WebGPU-to-WebGL fallback, a loader that paints in under a second, progressive loading, touch and performance on phones, hosting, link previews and publishing on OpenDrama.
The seven layers that take a Three.js scene from flat tech demo to a console-grade image: colour management, tone mapping, HDRI lighting, one shadowed key light, fog, bloom and lens, with the frame-time cost of each.
Tesana says a user built this fly-to-any-planet game from one prompt plus an image reference. We look at what that single prompt has to carry, and what it can't.
Tesana says Aestra's first build took one weekend. We break down how it makes permanent death harsh without driving players away.
jay88 had Claude Opus 5.5 build a Three.js pumpkin you carve with four tools and then light, in one Claude Code session. The prompt was two lines; the real work is the carve-map technique the model chose.
Sac had Opus 5.5 build a 33-second game teaser in Blender from one prompt: a rose, prop-topped planets, a fox and a sunset over clouds. How simple geometry gets real depth.
Shared by @threejs: eight Monet paintings turned into WebXR worlds you can walk, fly and paint in. How the brushstroke renderer, the hand controls and the voice guide work.
Sac's 36-second pre-launch scene reel from The Little Prince: the King's palace, a glass-roofed stone house, the Lamplighter's rock. Landmarks, motifs, a shared kit, lighting and per-planet bundles.
lxfater had a locally hosted ~100B-parameter model write the code, Tripo3D make the car shells and Godot hold it together: a Mini 4WD game with car select, swappable parts and races, in one day.
Sac has been driving AI video with Blender gray-box renders to cut rerolls. His side-by-side shows why: one prompt, two models, two different stagings. How the whitebox route works.
Stefan Vaskevich gets a deer and a boar walking in 3D with no mocap and no hand keyframing. The whole trick is one colour per leg; here's the pipeline and where it slips.
nicekate's 10-minute video covers Gemini 4 Argon's scores, pricing, demos and the five tips from Google's developer guide. We keep what matters to creators: what it costs, where it isn't first, and how the five tips apply to making games.
Stefan Vaskevich opens the local workspace he makes games in: chats beside project pages, saved skills, MCPs for the engines, and a 36-hour Unity run documented card by card.
A free tool copies a filmed person's motion onto any Mixamo character and hands back a skinned GLB. How to set it up in Blender, how long the first run takes, and where it stops.
A SIGGRAPH Asia 2026 method turns missing, extra and jittery mocap markers into clean body-and-hands motion, with one model for any layout. The code isn't out yet; here's what it means for game animation.
Stefan Vaskevich posted Gemini 4 Argon rebuilding Never Gonna Give You Up in 3D, beside the original. Thirty seconds show what the model gets right in 3D and animation, and what it doesn't.
Meng To used the open-source Anatomy skill to turn one prompt into an animated SVG explainer of Seedream's image layers. The prompt, the one follow-up, and how the skill works.
Meng To is building an icon-and-illustration-to-3D tool with Opus 5.5: presets, materials, bases and animated backdrops. How it is structured, and where an SVG-to-3D pipeline breaks.
