Strata Landscape

Math essence

Each layer i has frequency f_i, phase φ_i, hue h_i. The layer top/bot lines are y = i*L + amp(t) * sin(f_i * xs/4 + φ_i + 0.04*t). Per-step amp += perturbation_growth, capped at 0.6. The pixel mask (ys in [top, bot]) selects that layer's palette colour; a 0.85 blend over the previous canvas keeps strata readable.

Parameters

Reference

First 60 lines of reference.py. Full source: library_external/akten_quayola/StrataLandscape/reference.py

from __future__ import annotations

from typing import ClassVar

import numpy as np


class StrataLandscape:
    """Strata Landscape — Quayola, *Strata* series (2008–ongoing).

    Quayola's Strata series — begun with the Roman baroque churches in
    2008 — deconstructs classical paintings and architecture into
    horizontal layers of rock-like strata, each layer gaining its own
    degree of digital perturbation.  This formula builds that kernel
    directly:  N horizontal bands stacked vertically, each band
    displaced by a slowly mutating sinusoid and shaded with a layer-
    specific palette tone.

    As t advances, the perturbation amplitude grows and the layers
    drift sideways — a "geological" evolution through the work of
    erosion and algorithmic weathering that Quayola describes when
    discussing Strata's geological metaphor.
    """

    family: ClassVar[str] = "akten_quayola"
    sketch: ClassVar[str] = "StrataLandscape"

    def __init__(
        self,
        num_layers: int = 60,
        dim: int = 512,
        perturbation: float = 0.08,
        perturbation_growth: float = 0.0003,
        depth_curve: float = 1.4,
        seed: int | None = None,
        background: tuple[int, int, int] = (8, 8, 14),
        **_,
    ) -> None:
        self.num_layers = int(num_layers)
        self.dim = int(dim)
        self.perturbation = float(perturbation)
        self.perturbation_growth = float(perturbation_growth)
        self.depth_curve = float(depth_curve)
        self.bg_color = np.array(background, dtype=np.float32)
        self.reset(seed=seed)

    def reset(self, seed: int | None = None) -> None:
        self.rng = np.random.default_rng(seed)
        self.t = 0.0
        self.amp = self.perturbation
        self.freqs = self.rng.uniform(0.005, 0.02, size=self.num_layers).astype(np.float32)
        self.phases = self.rng.uniform(0.0, np.pi * 2.0, size=self.num_layers).astype(np.float32)
        hues = np.linspace(0.0, 1.0, self.num_layers, dtype=np.float32)
        self.hues = hues + self.rng.normal(scale=0.04, size=self.num_layers).astype(np.float32)
        self.hues = np.clip(self.hues, 0.0, 1.0)
        self.canvas = np.tile(self.bg_color, (self.dim, self.dim, 1))

    def step(self, n: int = 1) -> None:
        for _ in range(n):
            self.amp += self.perturbation_growth