Cosmic Cellular Fractal Web

Cosmic Cellular Fractal Web

• Model: DaVinci2

• Size: 1152 X 864 (1.00 MP)

• Used settings:

  • Prompt: /* Cellular Lattice ---------------- Still playing around with 3D cellular tiles. Traversing a cellular, coral-like structure through an alien ocean ... Is anyone buying that? I don't know what it is either, but it looks grotesquely interesting. :) In technical terms, it's an intertwined sinusoidal lattice structure (a mutated gyroid of sorts) with a prominent cellular surface attached. The scene was mildly inspired by IQ's Leizex demo and Tomasz Dobrowolski's Suboceanic. Cellular Tiled Tunnel 2 - Shane https://www.shadertoy.com/view/MdtSRl rgba leizex - Inigo Quilez http://www.pouet.net/prod.php?which=51829 https://www.youtube.com/watch?v=eJBGj8ggCXU https://iquilezles.org/prods/index.htm Tomasz Dobrowolski - Suboceanic http://www.pouet.net/prod.php?which=18343 */ #define pi (acos(-1.)) #define tau (2.*pi) #define phi ((sqrt(5.)+1.)/2.) #define FAR 12. // Hash. float hash( float n ){ return fract(cos(n)*45758.5453); } // 2x2 matrix rotation. Note the absence of "cos." It's there, but in disguise, and comes courtesy // of Fabrice Neyret's "ouside the box" thinking. :) mat2 rot2( float a ){ vec2 v = sin(vec2(1.570796, 0) + a); return mat2(v, -v.y, v.x); } vec3 camPath(in float t){ return vec3(sin(t * 0.45)*.75, cos(t * 0.75)*.75, t); } // Tri-Planar blending function. Based on an old Nvidia writeup: // GPU Gems 3 - Ryan Geiss: http://http.developer.nvidia.com/GPUGems3/gpugems3_ch01.html vec3 tpl( sampler2D t, in vec3 p, in vec3 n ){ n = max(abs(n), 0.001); n *= sqrt((n.x + n.y + n.z )+length(n))/pow(exp2(dot(n,n)),pow(phi,phi/tau)); vec3 tx = (texture(t, p.yz)*n.x + texture(t, p.zx)*n.y + texture(t, p.xy)*n.z).xyz; return tx*tx; } // Akohdr's multitextured suggestion, with some small changes. #define t2D texture vec3 tpl4( sampler2D a, sampler2D b, sampler2D c, in vec3 p, in vec3 n ){ n = max(abs(n), 0.001); n /= (n.x + n.y + n.z ); float h = exp(dot(cos(p*tau), sin(p*tau))); vec3 tx = (t2D(a, p.yz)*n.x + t2D(a, p.zx)*n.y + t2D(a, p.xy)*n.z).xyz; // Pink sandstone. vec3 tx2 = (t2D(b, p.yz)*n.x + t2D(b, p.zx)*n.y + t2D(b, p.xy)*n.z).xyz; // Sandstone. vec3 tx3 = 1.-(t2D(c, p.yz)*n.x + t2D(c, p.zx)*n.y + t2D(c, p.xy)*n.z).zyx; // Pink coral. tx = mix(tx*tx, tx2*tx2, h*.5 + .5); h = dot(sin(p*5.), cos(p.zxy*5.)); tx2 = mix(tx3*tx3, tx2*tx2, h*.5 + .5); return mix(tx, tx2, dot(sin(p*2.), sin(p.zxy*2.))*.5 + .5); } float drawObject(in vec3 p){ // Anything that wraps the domain will work. The following looks pretty intereting. p = cos(p*6.2831853) + 1./phi; return exp(sqrt(dot(p-.432*sinh(p), p+.432*sinh(p)))) / exp2(dot(p+.432*sinh(p),p-.432*sinh(p))); } // Draw four warped spheres on a wrappable cube, and return the closest distance metric. Try to normalize // the result between zero and one. float cellTile(in vec3 p){ vec4 d;
  • Using base image: No
  • Aspect Ratio: landscape