Zibra Liquids now has neural colliders. What are they? And what happened to the voxel ones?

Zibra Liquids offers a broad spectrum of features. Among them – colliders, which are essential while working with effects. Our solution has two types of liquid colliders – analytic and AI-powered ones. 

Analytic Colliders

Analytic methods allow you to add simple geometry, such as spheres, boxes, capsules, toruses, and cylinders as liquid colliders. In these colliders, collision is calculated with analytic functions, making them very fast. 

Zibra Neural Colliders

On the other hand, AI-powered are based on our custom technology of a neural SDF representation and enable adding of arbitrary geometry as a collider for liquid. 

With these colliders, there is no need to divide complex shapes into simpler ones. Liquid interaction is realized with any complex object as a whole and can be adjusted in a few clicks. 

For a long time, our cutting-edge AI-powered colliders were called Voxel colliders. You can find this name in our previous tutorials, blog posts, and even documentation. But starting from the 1.4.3 version of Zibra Liquids, we are changing their name to Zibra Neural Colliders.

Advanced Neural Network Architecture

Zibra Neural Colliders use a new, more advanced neural network architecture that is more performant. In fact, it is almost as good as sampling SDFs from a 3D texture and takes only about 1.8 nanoseconds per sample on an RTX 2060 mobile. 

They also allow for a greater level of compression while taking up to 300x less memory compared to standard 3D texture with the same quality. 

As for functionality, it did not change significantly, but there are a couple of improvements. For example, our new neural network handles high-detail objects much better, including those with really thin walls. 

With new neural network architecture, you can also add even more AI-powered liquid colliders into the scene. 

Overall, it grants a much more consistent SDF, quite similar to one computed directly from mesh, while being much faster and requiring less memory.

Check out the difference!