Illusions in Graphics

What's easier? Simulating reality or simulating the illusion of reality? What does it take to lie convincingly in graphics?

Illusions in Graphics We all know that graphics are an illusion.

What we spend most of the time doing in "graphics rendering" or "graphics programming" is just calculating how some tiny lights we call pixels respond and simulate reality.

Early processors and graphics pipelines were not powerful enough to simulate reality, and we had not studied enough how nature behaved in order to convincingly simulate it.

Much of graphics programming is about convincing the player to accept an artificial world as real.

Or at least that's what most games try to do, because there is some creative room.

For instance, we find ourselves with games like Celeste that are not trying to simulate reality, they try to create a sort of cozy, nostalgic scene.

The games that came first, probably did not want to simulate reality, because they could not.

They accounted for the imagination of the player, and they were not trying to create a photorealistic world but rather tell a story.

With the technologies we have today, we can have photorealistic graphics, but we are still not using them everywhere: they are expensive, and not always the best choice for the game we want to make.

Clair Obscur: Expedition 33 scene Celeste Clair Obscur was made with Unreal Engine 5, and Celeste was made with XNA, a framework for making games in C#.

Both games are very different, but they are both successful in their own way.

They are both trying to create an illusion, but they are doing it in different ways.

And I like them both, and both could be considered masterpieces in their own right.

But they are different visually, so: What's going on?

Graphics as a medium of expression Graphics not just "make your game look alright", they empower you to express your vision.

They are a medium of expression, and they can be used to create different moods, atmospheres, and feelings.

They can be used to tell a story, to evoke emotions, to create a sense of immersion, and to make the player feel something.

In this post, I would like to discuss the evolution of photorealism in graphics, and how it has influenced the way we create games.

I will also discuss how we can use graphics to create different illusions, and how we can use them to tell stories and evoke emotions.

Nostalgia is one of the powerful feelings that we can evoke with graphics, and it is one of the reasons why pixel art is still popular today.

It reminds us of the games we played when we were younger, and it evokes a sense of warmth and comfort.

It is not trying to simulate reality, but rather evoke emotions and tell a story.

On the other hand, photorealism is trying to simulate reality as closely as possible, and it is trying to create a sense of immersion and believability.

This totally depends on the story or gameplay you want to convey.

Traditionally, art style does not determine depth, but it influences how the experience is perceived and what cognitive expectations the player brings into the game.

One example would be Balatro.

But there are games like Hollow Knight that despite being 2D, the art style is new and it feels deep and thought after.

This is because games that tend to fit in our planet like GTA, Yakuza or The Last of Us must make the player feel like they are in a real world, and that is what photorealism is good for.

On the other hand, games that are more abstract or that are not trying to simulate reality can use different art styles, and they can use different illusions to create different moods and atmospheres.

The evolution of photorealism in graphics The early days of graphics Graphics have come a long way since the early days of gaming.

In the beginning, we had very limited hardware and software capabilities, and we had to make do with what we had.

We had to use pixel art, and we had to use simple shapes and colors to create our games.

We had to rely on the player's imagination to fill in the gaps, and we had to use clever tricks to create the illusion of depth and realism.

But they have always been an illusion, and they have always been a medium of expression.

In the early days of 3D gaming, graphics were mostly simple and blocky, and the lighting was not realistic at all.

But as technology advanced, we were able to create more complex and realistic graphics.

We were able to create more detailed models, more realistic lighting, and more complex shaders.

We were able to create more immersive worlds, and we were able to create more believable characters.

But we were still not able to create photorealistic graphics, and we were still not able to create the illusion of reality.

Lambert Diffuse Lighting This was probably the first revolution in graphics, and it was a big one.

It allowed us to create the illusion of depth and realism, and it allowed us to create more immersive worlds.

It was a simple model, but it was effective, and it was widely used in games for many years.

It basically tells: "The amount of light that hits a surface is proportional to the angle between the light and the surface normal." This is a simple model, but it was effective, and it was widely used in games for many years.

I = k{d} \max(0, \mathbf{L} \cdot \mathbf{N}) Lambert Diffuse Lighting Phong & Blinn-Phong Shading This was the second revolution and where most of hobby game engines start, including Atlas.

That's because Phong and Blinn-Phong shading models are simple and efficient, and they allow us to create more realistic lighting and shading effects.

They are based on the idea that the amount of light that hits a surface is proportional to the angle between the light and the surface normal, but they also take into account the angle between the light and the viewer.

This allows us to create more realistic specular highlights, which are the bright spots that appear on shiny surfaces when they reflect light.

So, they introduced the concept of specular highlights, which are the bright spots that appear on shiny surfaces when they reflect light.

This was a big step forward in terms of realism, and it allowed us to create more immersive worlds.

I = I{a} + I{d} + I{s} Where: I{a} is the constant ambient light, that allows us to simulate the light that is scattered in the environment and that is not coming from a specific source.

I{d} is the diffuse light, that allows us to simulate the light that is reflected in all directions from a surface.

It is based on the Lambert Diffuse Lighting model, and it is calculated as I{d} = k{d} \max(0, \mathbf{L} \cdot \mathbf{N}).

I{s} is the specular light, that allows us to simulate the light that is reflected in a specific direction from a surface.

It is based on the angle between the light and the viewer, and it is calculated as I{s} = k{s} \max(0, \mathbf{R} \cdot \mathbf{V})^{n} Blinn-Phong Shading Physically Based Rendering (PBR) This was the latest revolution in what we call "real-time graphics", and it is still evolving.

The next methods already push the limits of what simulation means.

What is special about the methods we have seen so far is that they are based on empirical models, which means that they are based on observations and experiments, and they are not based on the actual physics of light.

PBR, on the other hand, is based on the actual physics of light, and it allows us to create more realistic and accurate lighting and shading effects.

It is based on the idea that the way light interacts with a surface is determined by the physical properties of the surface, such as its roughness, its metallicity, and its albedo.

This allows us to create more realistic materials, and it allows us to create more immersive worlds.

This method introduced ideas like energy conservation and microfacet theory, which means that it takes into account the fact that light can be reflected, refracted, and absorbed by a surface, and it takes into account the fact that surfaces are made up of tiny facets that can reflect light in different directions.

This allows us to create more realistic materials, and it allows us to create more immersive worlds.

And this method gaves us what we call the rendering equation.

The equation that describes how light interacts with an object: L{o}(\mathbf{x}, \omega{o}) = L{e}(\mathbf{x}, \omega{o}) + \int{\Omega} f{r}(\mathbf{x}, \omega{i}, \omega{o}) L{i}(\mathbf{x}, \omega{i}) (\mathbf{n} \cdot \omega{i}) d\omega{i} PBR is a method that allows us to create more realistic and accurate lighting and shading effects.

It is based on the actual physics of light, and it allows us to create more realistic materials, and it allows us to create more immersive worlds.") From illusion to reality The use of Ray Tracing Ray Tracing is well known because it is slow but incredibly accurate.

It is based on the idea of tracing the light path as it goes through the scene.

Ray Tracing is normally confused with Path Tracing, but they are not the same thing.

Path Tracing uses Ray Tracing, but it serves a different purpose.

Ray Tracing is the interface that can be used in a bunch of different techniques, like: Screen-Space Global Illumination (SSGI): Simulating indirect lighting by tracing rays in screen space.

Ray-Traced Reflections: Simulating accurate reflections by tracing rays from the camera to the scene.

Ray-Traced Shadows: Simulating accurate shadows by tracing rays from the light source to the scene.

Path Tracing Path Tracing is the most accurate method we have for simulating light, and it is based on the idea of tracing the light path as it goes through the scene, but it also takes into account the fact that light can be reflected, refracted, and absorbed by a surface.

It is based on the rendering equation, and it allows us to create more realistic and accurate lighting and shading effects.

It is still very slow, but it is the most accurate method we have for simulating light, and it is used in some high-end rendering engines like Atlas Engine.

Path Tracing is what we call a "Monte Carlo integrator" built on top of Ray Tracing.

Path Tracing The future of graphics We are now seeing a growing integration of AI-based reconstruction techniques in modern graphics pipelines.

Technologies such as DLSS, FSR, and MetalFX allow games to render at lower internal resolutions while reconstructing higher-resolution images through learned or algorithmic upscaling.

This shift is fundamentally about perception rather than raw computation.

Instead of rendering every pixel at full cost, modern systems prioritize where detail is actually perceived by the human eye, trading physical accuracy for visual plausibility.

As a result, games can maintain higher frame rates or render more complex scenes while still presenting a high-resolution image.

In some cases, additional techniques such as frame generation further increase perceived smoothness.

While there has been debate around artifacts and reconstruction quality, these tools are not replacing artistic intent.

Instead, they expand the design space available to developers, allowing more complex scenes, richer lighting, and higher fidelity worlds within practical performance limits.

AI-assisted rendering is therefore becoming another layer in the same idea that drives all graphics: constructing convincing visual experiences under constraints.

In that sense, it is not a break from traditional rendering, but a continuation of the same goal.

Conclusion Think one second about the image that was in the cover of the post.

It was this one: Fluid Simulation Even this is tricking us since it is probably SPH, which is a model that renders small particles instead of a whole fluid, because nature is hard to simulate.

Nature is extraordinarily complex, and our simulations are always approximations.

We can only try to simulate it, and we can only try to create the illusion of reality.

But that is what makes graphics so interesting, and that is what makes it such a powerful medium of expression.

We can create different illusions, and we can create different worlds, and we can create different stories.

And that is what makes graphics so special, and that is what makes it such a powerful tool for game design.

Everything we see is a lie, but a convincing one.

And that is what makes graphics so fascinating, and that is what makes it such a powerful medium of expression.

Graphics is the same as painting, selling cars, writing a book or being a magician.

It is all about creating illusions, and it is all about creating a sense of wonder and amazement.

And that is what makes graphics so special, and that is what makes it such a powerful tool for game design.