One prompt, $17.57, 182 minutes: Claude autonomously simulates and rides a lightning bolt
TAbrodi · x · 2026-11-08
camperaigg's Fable 5.5 test: with a single prompt and zero code edits, Claude (Fable 5.5 Xhigh in Claude Code) spent 182 minutes and $17.57 in API costs to autonomously simulate lightning — a 30.5M-cell storm grid with rising air, rain, ice and charge (63 flashes logged), a dielectric-breakdown sim growing the bolt at 16 m grid resolution (-110 MV at the root), a camera riding the leader tip for all 60 steps, a 2.4M-blade prairie scarred with the same physics, and fully synthesized sound including 130 ms of silence before the crack. Reposter TAbrodi was impressed.
More from coding & agent
- Emergent UI Downside: Colleagues Now Expect You to Read Every AI-Generated Interface — sloppenheimer · 2026-11-08
- Dev Open-Sources a Three Kingdoms SLG Where Claude Plays Everything via MCP — Regular_Bison_2175 · 2026-11-08
- Meta Muse mobile review: persistent VM, editable memory files, and async approvals make it a control plane — Acceptable-Web-6296 · 2026-11-08
- Interp researcher: AI coders dismiss accumulated engineering norms at their peril — voooooogel · 2026-11-08
- Full Grok thermal printer video hits YouTube, edited with Codex — billyjhowell · 2026-11-08
- Recreating Seedance-Style AI Video with Blender Blockouts + MiniMax H3 Ref2Vid — Time-Ad-7720 · 2026-11-08