Photoreal rendering describes the goal of 3D graphics to produce an image that could plausibly be a photograph. It is less a single technique than an outcome defined by a set of physical accuracy criteria. These criteria include how light scatters across a surface, how shadows fall and soften with distance from their caster, how transparent materials refract the scene behind them, and how a material's finish responds differently to direct versus ambient light.
The term is widely used across industries that depend on visual accuracy for commercial or technical purposes: product visualization, automotive design, architecture, film visual effects, and e-commerce. In each context, photorealistic rendering serves the same purpose: it allows viewers to form accurate expectations about how something will look in reality before they have seen it in person.
Achieving photorealism requires that a rendering system simulate physical phenomena with enough fidelity that the human visual system cannot identify the output as computer-generated. Human observers readily perceive incorrect shadow softness, missing ambient occlusion, implausible material reflectance, or color temperatures that don't match the stated lighting environment.
Light in the physical world follows principles described by the rendering equation introduced by James Kajiya in 1986. The rendering equation expresses the total light leaving a surface point as the sum of light emitted by the surface and light reflected from all other directions, weighted by the surface's reflectance properties.
Ray tracing is the rendering technique most directly grounded in the physics of light. A ray tracer traces light rays from the camera through each pixel into the scene, computing interactions with surfaces and sampling light sources to determine shadows. Path tracing extends this by stochastically sampling multiple light paths per pixel, producing globally illuminated images with realistic indirect light. Offline rendering systems use path tracing with thousands of samples per pixel, running for minutes or hours. Real-time rendering uses rasterization with PBR approximations; hardware-accelerated ray tracing via NVIDIA RTX and DirectX Raytracing (DXR) has enabled physically-based real-time ray tracing in production.
Physically Based Rendering (PBR) is the material and shading framework that enables consistently photoreal results. PBR defines material properties in physically meaningful terms: how much light a surface reflects specularly versus diffusely, how rough or smooth the microsurface structure is, and whether the material is a metal or a dielectric. Materials defined using PBR parameters behave correctly across any lighting environment, a prerequisite for photoreal results in production contexts where lighting conditions are not fixed.
The glTF 2.0 format encodes material properties using the PBR metallic-roughness model defined by the Khronos Group, ensuring materials authored in tools like Adobe Substance 3D render consistently in any glTF-compatible viewer.
Photoreal rendering quality has historically been confined to contexts where rendering time and hardware cost are not constraints. Two converging developments are changing this. First, physically based shading is now standard in real-time rendering, meaning web-based 3D experiences can use the same material definitions as offline renders. Second, streaming architectures enable high-fidelity assets to be delivered progressively, so the quality ceiling for web-based 3D is no longer determined by download size. A photoreal product model can stream to a mobile browser and render with physically based shading, producing an experience that closely approximates a dedicated visualization tool.
3D physically based rendering (PBR) — The material and shading framework that underpins physically accurate, photoreal results.
Asset optimization — Preparing high-fidelity photoreal assets for efficient streaming and real-time delivery.
Gaussian splatting — A 3D representation method that captures photorealistic scene appearance from real-world photographs.
Streaming video reshaped media consumption; we're doing the same for 3D. Join a small team solving tough spatial streaming, content delivery, and developer experience challenges.
Technical deep-dives on adaptive spatial streaming, the infrastructure challenges of 3D at scale, and what we're learning as we build. Written by the team doing the work.
