Fast Starnest
This port replace GPU branching with ternary operator so it run faster.
Use it on a ColorRect Node.
Shader code
shader_type canvas_item;
uniform int iterations = 20;
uniform float formuparam = 1.00;
uniform int volsteps = 20;
uniform float stepsize = 0.1;
uniform float zoom = 0.800;
uniform float tile = 0.5;
uniform float speed = 0.001;
uniform float brightness = 0.002;
uniform float darkmatter = 0.100;
uniform float distfading = 0.650;
uniform float saturation = 0.750;
float SCurve (float value) {
float v = value;
value = v < 0.5 ? value : value - 1.0;
value = value * value * value * value * value * 16.0;
return v < 0.5 ? value : value + 1.0;
}
void fragment() {
vec2 iResolution = 1.0 / SCREEN_PIXEL_SIZE;
//get coords and direction
vec2 uv =FRAGCOORD.xy / iResolution.xy;
uv.y *= iResolution.y / iResolution.x;
vec3 dir = vec3(uv * zoom, 1.0);
float time = TIME * speed;
vec3 from = vec3(1.0, 0.5, 0.5);
from -= vec3(0.0, time, 0.0);
//volumetric rendering
float s = 0.1, fade = 1.0;
vec3 v = vec3(0.0);
for (int r=0; r<volsteps; r++) {
vec3 p = fma(vec3(s), dir*0.5, from);
p = abs(vec3(tile) - mod(p,vec3(tile*2.0))); // tiling fold
float pa, a = pa = 0.0;
for (int i=0; i<iterations; i++) {
p = abs(p) / dot(p,p) - formuparam; // the magic formula
a += abs(length(p) - pa); // absolute sum of average change
pa = length(p);
}
float dm = max(0.0, fma(-a, a * 0.001, darkmatter)); //dark matter
a *= a; // add contrast
fade = r > 6 ? fade * (1.0-dm) : fade; // dark matter, don't render near
v += fade;
v = fma(vec3(s,s*s,s*s*s*s), vec3(fade*a*brightness), v); // coloring based on distance
fade *= distfading; // distance fading
s += stepsize;
}
v=mix(vec3(length(v)),v,saturation); //color adjust
vec4 C = vec4(v*.01,1.);
C.r = pow(C.r, 0.35);
C.g = pow(C.g, 0.36);
C.b = pow(C.b, 0.38);
vec4 L = C;
COLOR.r = mix(L.r, SCurve(C.r), 0.7);
COLOR.g = mix(L.g, SCurve(C.g), 1.0);
COLOR.b = mix(L.b, SCurve(C.b), 0.2);
}



