forge-screen-space-reflections

To add screen-space reflections in SDL3 GPU, ray march through the G-buffer using depth, normals, and world-space positions to generate on-screen reflections without environment probes.

Ray marching in deferred rendering traces rays through screen-space depth and normal data to find intersections, producing reflections that are composited into the final scene.

Yes, screen-space reflections require a complete G-buffer containing world-space positions, normals, and depth to accurately calculate reflection vectors and intersections.

Yes, ray marched screen-space reflections are suitable for interactive graphics like wet surfaces, polished floors, and mirrors, integrating with existing lighting and shadow passes.

SSR uses ray marching through existing G-buffer depth and normal data to produce plausible reflective surfaces, avoiding the expensive setup required by environment probes.

Screen-space reflections only capture geometry visible in the G-buffer, meaning off-screen objects or surfaces occluded from the camera will not appear in the reflection composite.