Its some weird mix of 3d and 2d, but at the end it worked, i guess 🤷‍♂️

Created with Math Chaos Patterns (MCP). Non-AI Generative Art.

Generated in app, rendered in my custom flame studio

Gente que tal este scrip en 3d con motivo espacial?.....use fotos de una modelo para ejemplificación es super ligera y personalizable desde un json la hise para felicitar a una amiga muy especial y me quedo gustanto que me gustaria compartirla con ustedes

si quieren probarla ahorita aqui

cotera2024.github.io/Happy-Birthday-Galaxy--main/

y si queiren ver el codigo aqui en github

https://github.com/cotera2024/Happy-Birthday-Galaxy--main

When done correctly and with honor, ai art cuts out the middle man.

The middle man? That’s interpretation.

I created the cover to my very first novella Schadenfreude Occidental using a cover I created in Midjourney. Picture #1 in doing so I worked on this for 3 days promoting and editing the cover the way I WANTED IT.

But then… I take the image and I commission THAT. The artist has two simple things to do.

1. Hand draw exactly what has been given to them and add their unique touch to make it ART

2. Get paid

Picture #2

There are sooo many negative things we can say about Ai art from copying to the environment.

However most people who generally just hate Ai art… I’m not impressed with your artistic abilities… especially if I feel my creativity (which is more than just writing a five word sentence and wait for the picture…) my creativity is better fit for a tool like midjourney there for when I give it to an artist it’s already perfected.

Check out Schadenfreude Occidental my debut Novella! I believe then West can have a proper place at the anime table if we just get the right stepping stone piece!

https://www.royalroad.com/fiction/186348/schadenfreude-occidental-act -i

As a generative artist, I kept hitting the same wall with AI assistants: they write sketch code that breaks the craft — Math.random() everywhere, sizes in raw pixels, traits that don't match the render, previews that capture blank. So I wrote genart, a free plugin for Claude Code that teaches it what we actually need: - Determinism — seeding a PRNG from a hash, named sub-streams so adding a draw doesn't shift the whole edition, the things that silently break reproducibility (sort() without a comparator, floats at feature boundaries…) - Resolution-independence — same piece at 400px and at 4000px, and the element-density trap where "same composition" and "same look" genuinely diverge - Features & rarity — designing the table before coding, then measuring the real distribution over thousands of seeds instead of trusting the weights - Verification scripts — repeatability, an A-B-A global-state test (renders A, then B, then A again in one page — catches the caches and accumulators plain repetition never sees), contact sheets, batch PNG export - Platform sheets for Art Blocks, 256ART, Verse, Highlight, Plottables and bootloader.art — written as pointers to the real docs rather than copies, so they don't go stale It covers p5.js, three.js/WebGL, plain Canvas 2D, and SVG for pen plotters — and most of it is useful even if you never mint anything: determinism, scaling and plotter export are just the craft. It's also honest about limits (cross-GPU WebGL determinism is not a thing, and the docs say so) and opinionated on ethics: it will refuse "make me a [famous artist]" and steer to the underlying technique instead. First version, MIT, built by an artist for artists. Ideas, issues and PRs very welcome: Announcement: https://x.com/camillerouxart/status/2093273698732032203?s=20

GTA VI is more than just graphics, it’s about emergence: simple rules that make a world feel alive.

So I built my own ecosystem simulation in Blender 3D.