Wednesday, August 20, 2025

Evolving attractors Part3

In a previous show, I created a visual using Thomas' cyclically symmetric attractor, but for the next performance, I wanted to revisit and expand on the idea. My goal was to explore new attractors, tweak existing ones, and experiment with their behavior.

I went back to Thomas' attractor and modified the shader that generates position instancing data. By adjusting different parts of the differential equations, I eventually stumbled upon a shape that I found really interesting.

// Example Compute Shader

uniform float b_val;

layout (local_size_x = 8, local_size_y = 8) in;
void main()
{
	vec4 color = texelFetch(sTD2DInputs[0], ivec2(gl_GlobalInvocationID.xy), 0);

	vec3 particale_pos = color.rgb;

	vec3 pos = particale_pos;
	

	float b = b_val;
	vec4 final_color;
	/* dx */final_color.r= sin(pos.y) - b*pos.x;
	/* dy */final_color.g= 0.9*sin(pos.z*2.0) - b*pos.y;
	/* dz */final_color.b= sin(pos.x) - b*pos.z;
	/* life */final_color.a=1.0;

	vec4 imageOut = final_color;
	imageStore(mTDComputeOutputs[0], ivec2(gl_GlobalInvocationID.xy), TDOutputSwizzle(imageOut));
}

By simply adding some extra constants, I was able to generate this new attractor shape:


To enhance the depth, I added a second attractor:


To complete the visual, I brought back the original Thomas attractor:

This was a continuation of my previous visual world, where I had used the Thomas attractor along with GLSL compute shaders to add distortions and color variations to each instance within the attractor.

As before, I pre-programmed three b values that I could switch between live using my MIDI keyboard.

b=0.2081

and the two new ones being

b=0.29818




and b=0.43018




No comments:

Post a Comment

Stage 1 Initial Block Out - Layout - 'F' Stage

For my block out I procedurally made a chandelier , and I ran into a few challenges along the way. First, I had to figure out how to make a ...