so once again a growth algorithm. this time it is about triangles. I started the simulation with three base pyramides. there is a variable chance of growing another triangle on the current one - here it was 50 %. intersections between the triangles are avoided.
after growing a defined amout of iterations all surfaces are transformed into frames. the size of the frame-openings are linked to the z-position of each single surface.
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiYvOVtywW1eoZXW7rSfxCMYakU7WpJJ3iA-0NXnB-B10IYPkrINp6kC8VVFuoieG_Gs73Rh9ZwMefQ99cj8_nAJEkFcm2UOkan38bjlvKzktIqcMvcldDlcbJ8SKMOKmC336r0FiW-xHU/s400/01_triangle+growth+5.jpg)
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgljOGQUmGYlU-2HIhnefH_U15y12kljyKtLqcsKxEO84TS4-3CdjdWGTHuk5EutmBfBi5JYk-n-kEWifFApOahku6cQ7nT4wlsdIeaHQRlA9szasV4mT4UUhM0iI_11lkW950dtrJ3DrY/s400/01_triangle+growth+7.jpg)
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjH4fCGiTx2pO4iFwwyjSHDkwDA0WqkxIFubYjfnTx7itq7BwUW5jgpCAK43STX6X2AwlgBUnJGSt9gpLntBhS4QMKIYHOe738B38BAPGk8N1F2kXcsJe-_ElTpIxCLbVIGRyQYswETHcU/s400/01_triangle+growth+6.jpg)