At some point in the last few years, a default aesthetic emerged for high-end 3D websites.
Dark background — usually very dark blue, almost black. Floating geometric object in the center of the scene, often a sphere or an abstract form. Bloom effect making edges glow. Particle field suggesting depth and scale. Sans-serif display type in white or light grey. Subtle camera drift.
This aesthetic appears everywhere now. Agency portfolios, product launch sites, AI company homepages, creative developer demos. The technical execution varies — some are more sophisticated than others — but the visual language is nearly identical.
And it's not because this aesthetic is uniquely beautiful or effective. It's because it became the aesthetic that Three.js tutorials produce. It's what you get when you follow the path of least resistance through the available tools with no strong creative direction of your own.
How Default Aesthetics Form
Every medium develops defaults — the visual language that emerges when practitioners work with the tools without questioning the tools' built-in assumptions.
In the early days of the web, the default was light grey backgrounds, Times New Roman body text, and underlined blue links. The tools assumed those choices, so that's what everything looked like.
When CSS made custom typography accessible, everything shifted to a smaller set of web-safe fonts. Then Google Fonts arrived and everything shifted to Roboto, Lato, and Open Sans.
Right now, Three.js with bloom and a dark void background is the path of least resistance in 3D web development. The tutorials show you how to set up a scene with those parameters. The examples in the documentation use similar configurations. When you follow tutorials to build your first scenes, you learn to make things that look like the tutorials.
That's not a criticism of tutorials. It's just how defaults propagate. The problem is when practitioners accept the default as a creative choice rather than a starting point.
The Commitment Problem
Strong visual identity requires commitment to a specific world — a specific palette, a specific material language, a specific atmospheric quality that is maintained rigorously throughout the experience.
Commitment is uncomfortable, especially early in a project. When you commit to something unusual, you've made a decision that rules out many other possibilities. What if the client doesn't like it? What if it reads as too niche? What if the unusual thing turns out to be wrong?
The safe choice is to stay close to the established aesthetic. Dark background. Glow effects. Recognizable components. The visitor has seen this before and knows how to relate to it.
But "safe" and "memorable" are almost never the same thing. The experiences that get passed around, that come up in conversation months or years after they were built, that genuinely differentiate a brand — these are invariably committed to a specific visual world that is not the default.
Bruno Simon's portfolio is a game you drive a car through. Nothing about that is the obvious approach for a developer portfolio. It works precisely because of that commitment.
The One-World Rule
One principle that separates visually distinctive experiences from generic ones: every element in the scene should be explainable by the same world.
If your world is deep ocean — bioluminescent, vast, slow-moving — then every decision in the experience should come from that world. The color palette comes from it. The scale of things comes from it. The pacing of the camera comes from it. The way light behaves comes from it. Even the typography should feel like it belongs to the same sensibility.
When experiences look generic, it's usually because they're mixing worlds without knowing it. The main object comes from one aesthetic. The particle system comes from another. The typography comes from a completely separate set of references. The result is visually incoherent — each element is reasonably well-executed in isolation, but they don't add up to a place.
A coherent world doesn't require unusual choices. It requires that all the choices come from the same source. Medieval stone and warm candlelight is a world. Sterile white and precise geometric forms is a world. Warm amber and soft organic curves is a world. Any of these can be executed with the same Three.js toolkit that produces the default dark-sphere-with-bloom aesthetic.
The choice of world happens in creative direction, before any technical work begins. Most projects skip that step, which is why most projects end up at the same default.
When Dark and Glow Is the Right Choice
None of this is to say that dark atmospheric scenes are wrong. The dark/void/bioluminescent aesthetic is genuinely beautiful and appropriate for a real range of projects — especially anything that wants to communicate depth, mystery, the unknown, or something that lives in the dark and glows from within.
The problem is not the aesthetic. The problem is the aesthetic without a reason.
When a dark atmospheric scene comes from a specific creative brief — a specific story about a specific world that has been thought through before a line of code was written — it can be extraordinary. When it comes from "this is what Three.js scenes look like," it's just another one.
The test is simple: can you explain every visual choice by reference to the world you're building? If yes, the aesthetic is working. If the honest answer is "it looks cool" or "it's what the tutorial showed" — that's the point where the creative direction conversation needs to happen.
Escaping the Default
The fastest way to develop visual distinctiveness is to stop looking at other web experiences when you're trying to generate creative direction.
The default aesthetic spreads precisely because creative developers look at what other creative developers are making and take those existing aesthetics as the available options. If everything you're consuming as reference is dark-sphere-with-bloom, you'll have a very hard time imagining anything else.
The references that produce distinctive work tend to come from outside the medium: architecture photography, museum exhibition design, fashion editorial, experimental film, scientific visualization. These fields have developed visual languages that haven't been filtered through web development assumptions, which means they contain possibilities that don't exist in the default web aesthetic.
The visual language of a deep-sea research vessel is not the same as the default Three.js aesthetic, even though both are dark and involve the ocean.
The visual language of a brutalist concrete archive is not the same as the default web design aesthetic, even though both use dark typography on light backgrounds.
Those differences are the thing worth pursuing. Not to be different for its own sake — but because a specific visual language communicates something that a generic one cannot.
A website that looks like itself is a website people remember.