Learn core principles of the Arnold render engine.
- [Instructor] Let's take a look at some core concepts in Arnold rendering. Arnold is a physically-base renderer. It simulates the physics of lighting and materials with real-world physical properties. The parameters and work flows of physically-based rendering are slightly different from the legacy shading models that are probably familiar to you. With these new methods, we get a more accurate simulation of materials and lighting. And because Arnold is designed for simplicity and ease of use, we achieve photo realism with very little effort.
To use Arnold most effectively, we should know a little bit about the algorithm that does the computations. Arnold is what's known as a brute force renderer, and in fact it's named after a famous bodybuilder and politician Arnold Schwarzenegger. The name is a nod to the power of brute force approach, and in rendering, brute force simply means that the global illumination is calculated on every pixel of the frame. Without optimization, a brute force rendering is more computationally expensive than GI sub-sampling techniques, such as final gathering or irradiance mapping.
Those methods don't calculate GI on every pixel. Brute force has to do more work sampling every pixel, lengthening render times. But the genius of the Arnold renderer is that it's an optimized brute force algorithm. Arnold may be a brute but it's also very smart. It's much faster than similar renderers, such as Autodesk RayTracer, or ART. Arnold is an unbiased renderer. This means that no statistical bias has been introduced into the render calculations in order to speed up render times.
By contrast, mental ray is a biased renderer, and the distinction is in some ways a philosophical one because either biased or unbiased renderers can achieve a photo real result. But Arnold has taken the path of unbiased. It's more physically accurate, but it some situations that may come at a cost of longer render times. Arnold falls into a specific category of renderers called a Monte Carlo path tracer. This is a method in which rays are randomly fired out from each surface.
That's what gives Arnold and other Monte Carlo renderers their characteristic grainy look. It also facilitates fast, interactive performance in the Arnold render view. Energy is conserved by default in Arnold. In other words, it's an accurate simulation of how light travels, bounces, and dissipates in a scene. To optimize that simulation, we have lots of control over the accuracy of the computations. Either globally or through each individual light or material, we can control the sampling, which is the fidelity and also the ray depth, which is the number of bounces.
We can easily dial these controls up or down to control the look of the shot and to optimize render times. Each one of the components in the simulations, such as diffuse bounces or transmissive refractions, can be optimized separately. Although Arnold defaults to an energy-conserving lighting simulation, we have some artistic freedom to break the rules. For example, we can adjust the amount of bounce light or color bleed on a material to achieve non-physical results.
And those are some key concepts in Arnold rendering.
- Arnold rendering concepts
- Lighting with Maya and Arnold lights
- Controlling exposure
- Filtering light with Gobo
- Light attenuation with Decay
- Image-based lighting with Skydome
- Exterior daylight with Physical Sky
- Arnold Standard Surface material attributes
- Mapping material attributes
- Rendering refractions
- Mesh subdivision and displacement at render time
- Shading effects such as ambient occlusion and vertex color
- Camera effects such as fisheye and depth of field
- Animation image sequence rendering