Render / Rendering

What does rendering do?

Rendering is the computational process of turning a 3D model into a 2D photorealistic or stylized image. This article covers computer graphics rendering only, not external wall plaster coatings. The term in Slovak, Vykreslenie (render), refers exclusively to digital graphics in architectural practice.

At its core, rendering simulates how light leaves a source, bounces off surfaces according to their material properties, and reaches a virtual camera positioned in the 3D space. The renderer must solve a complex physics problem: given the geometry of surfaces, their reflectivity and roughness, the position and color of light sources, and the camera's position and lens properties, what light intensity and color arrives at each pixel? Modern rendering engines solve this by tracing paths of light rays through the scene, accounting for reflection, refraction, diffuse scattering, and shadow.

Unlike architectural visualization, which focuses on the deliverable product and client communication, rendering is the underlying technical computation. A visualization might use several different rendering engines or techniques; rendering is what those engines actually do.

What rendering algorithms exist for architectural work?

Architects and visualization studios choose rendering algorithms based on speed, accuracy, and project stage. The main approaches differ in how they trace light through the scene.

Rasterization projects geometry onto the screen and uses pre-baked lighting and textures; it is the foundation of real-time graphics and game engines. It cannot easily simulate complex light behavior such as caustics or accurate subsurface scattering, but it is fast enough for interactive design review.

Ray tracing traces rays from the camera into the scene and bounces them once or a few times off surfaces. It produces accurate reflections and sharp shadows but struggles with diffuse interreflection (light bouncing many times between dull surfaces). Rendering an image typically takes minutes to hours at high quality.

Path tracing traces rays from the camera and bounces them many times before accumulating the final color. Each pixel starts as random noise; as the renderer shoots more samples per pixel, the image becomes progressively cleaner and more accurate. This resolves diffuse lighting, soft shadows, and global illumination: the subtle interplay of light between surfaces that makes a space feel real. Path tracing is the physically-based standard for still architectural images and is now GPU-accelerated in modern engines, reducing wait time from hours to minutes or seconds for typical scenes.

What determines whether a render looks convincing?

Polygon count and shading alone do not make a photorealistic image. The actual ingredients are simpler to name but harder to execute.

A minimalist exterior with exceptional lighting and materials often outperforms a highly detailed model under dull light. Lighting and materials are where an architect can most effectively intervene to improve a render.

How does sampling determine render quality?

In path tracing, quality is bought with time. Each ray traced through the scene costs computational work. More rays per pixel mean less noise, but also longer render time. A typical workflow renders at 256 or 512 samples per pixel, producing a nearly clean image in 10–30 minutes on a modern GPU. Photorealistic stills for publication often use 1024–4096 samples and wait 1–4 hours.

Denoising algorithms (such as OptiX Denoiser) analyze the noisy render and remove grain while preserving fine details. A denoised 128-sample image often looks nearly as good as a 1024-sample image without denoising, saving substantial time. However, denoising trades a small amount of fidelity: fine specular highlights or distant textures may blur slightly, in exchange for speed. Professional workflows routinely denoise to balance quality and turnaround.

Render time is predictable: it scales roughly linearly with sample count and scene complexity. A complex architectural interior with many bounces may take twice as long as a simple exterior. GPU hardware (RTX 40-series, A100, etc.) is essential for practical turnaround times; CPU rendering is now rarely used for production work.

What is the ethical responsibility of architectural rendering?

A photorealistic render is a compelling and persuasive document. Clients and planning authorities trust it as a preview of reality. There is a real ethical line between showing a design compellingly and misrepresenting it.

Renders that show impossible lighting (perpetual golden hour, physically impossible shadow angles), vegetation that will take 20–30 years to mature, or material finishes the budget will not afford are ethically problematic. A client standing in front of the finished building in normal daylight should recognize it without surprise. The principle is straightforward: if a render shows something that the design will not actually deliver under normal conditions, it is misleading.

This is especially important in Slovakia's context, where the gap between render and reality can directly affect clients' trust in architects and their willingness to invest in quality design and specification.

Why is a render only a snapshot?

A single render captures one moment: one time of day, one weather condition, one viewpoint, one season. A building looks different under noon sun, sunset, overcast skies, and night lighting. The same interior appears different in summer afternoon light versus a rainy morning.

When the point of communication is the core design idea rather than material finish, architects often present stylized or diagrammatic imagery instead: sketch renders, wireframes, or abstract color studies to avoid the false precision of a single photorealistic snapshot. For client presentations focused on material and finish, multiple renders from different times of day and seasons are more honest. For planning submissions, exterior renders in neutral daylight typically suffice, but context and surroundings must be accurate.

A render is a tool for communication and decision-making, not a promise of a single, unchanging appearance. Skilled architects use this limitation intentionally, choosing lighting and composition that support the design narrative without making false claims about how the building will always look.