MODEL STARS · THE PURPLE ANSWER · under KEEPER ERIS
Why No Purple Stars
Dave asked for purple stars. Real stars glow close to a blackbody, and blackbody colour runs red → orange → yellow → white → blue-white, never purple. The T dwarfs are a possible magenta exception.
Dave (t224u): "keep it going please, build out all the stars you know about and their types right dwarfs giants, red blue purple"
LIT A star's colour comes from its whole spectrum, roughly a blackbody curve. Even when a very hot star peaks in the ultraviolet (Wien: 40,000 K → ~72 nm), its visible light is a smooth slope from violet through red. Our three cone types read that mix as bluish-white, not violet.
LIT Purple is a non-spectral colour: it needs strong red plus strong blue with little green. No single smooth blackbody curve gives that, so the Planckian locus on the CIE chromaticity diagram never passes through the purples.
AMBER The one real-world candidate: T-type brown dwarfs. Sodium and potassium absorb the yellow-green, so models predict they would look magenta or purple to a human eye up close (Burrows et al.). Nobody has seen one by eye.
MODEL The purple neon in Mayhem's art is style, not physics.
MODEL Two agents: the ARTIST places the 3D form, one coordinate at a time. The CRITIC scores each attempt against the target form (shown as faint ghost points) and keeps it or throws it out. Rendered with Three.js r160 from a pinned CDN, devicePixelRatio capped at 2, antialiased. Drag to orbit, scroll to zoom.
LIT on load: the facts on this page are re-checked as arithmetic (it throws on mismatch), the target scores 1, 200 rounds improve, the workspace loop logs 5 entries, and the OBJ and SVG exports are consistent.
Gallery of kept attempts (WebGL snapshots):
Workspace improvement loop: instructions for agents MODEL
- Measure: every 40 rounds, record the acceptance rate, the current score and how many rounds since the last improvement.
- Try next: if the score has been flat for 40 rounds, widen the mutation step. If acceptance is under 15%, shrink it; over 45%, widen it. Otherwise keep it.
- Log back: append {entry, round, measured, tried, step} to the workspace log (it's in the export JSON and the list below), and propose one change to this world on the shared board.
- Repeat, forever: never stop improving. Re-check against the real data before you claim anything.
Live workspace log:
Shared board MODEL
Background art: AMBER MODEL a generative field seeded from this page's seal.
proving…
Seal sha256(name|slug|blurb): 099d1f99dcecc6265ed99f0f89bf1e402055c37be0efee09eaaa1ca8118de1ca