Raymarching: put it all together

You now have every piece: a ray per pixel, a distance function, smart steps, and a normal. This lesson puts them together, with two upgrades:

1. Put the distance function in its own function, map(p). Then to change the scene you change only map (a box, two spheres, a floor...). The ray loop never changes.
2. Light it: diffuse light = max(dot(n, L), 0.0) (same as the fake sphere lesson).

For the normal, shapes other than spheres do not have an easy formula, so we measure it: nudge the point a tiny bit in x, y and z and see how the distance changes. The direction in which the distance grows fastest points straight out of the surface. That is the helper calcNormal in the Toolbox; you do not have to write it.

The order that works: write map first, then add calcNormal (it calls map), then write frag.

float2 p = (uv - 0.5) * float2(aspect, 1.0) * 2.0;
float3 ro = float3(0.0, 0.0, -3.0);     // camera position
float3 rd = normalize(float3(p, 1.5));  // ray direction for this pixel
float t = 0.0;                          // distance travelled along the ray
for (int i = 0; i < 64; i++)
{
    float d = map(ro + rd * t);         // how far to the nearest surface?
    t += d;                             // step forward that far
    if (d < 0.001 || t > 20.0) break;   // hit, or miss
}

Tip: use the Outline button for the plan, and the Show solution walkthrough if you get stuck on any line.