The Breda framework

Named after our hometown, the Breda framework is an extensive collection of sophisticated tools and applications aimed at advanced computational and graphical tasks, developed by experts in the field. It includes a variety of functionalities such as acceleration structure building, neural network training and inference, 3D graphics processing, and ray tracing, offering a comprehensive solution for asset management and specialized needs like NeRF viewing and acceleration structure generation.

Raytraced dynamic global illumination

Available under license. By leveraging hardware accelerated ray tracing we're able to deliver real-time dynamic global illumination with minimal artist intervention. No proxy geometry, no manually setting "bounce color". Instead we'll evaluate material properties as needed in a frame budget that can fit a mobile power envelope.

Machine learning framework

With our machine learning framework tailored for game developers we can deliver recent state of the art rendering and machine learning techniques.

Real-time path tracers

Our team of rendering experts has developed a versatile real-time path tracing implementation, designed to meet the diverse needs of various applications. This innovative solution is implemented in two distinct forms: as a wavefront path tracer and as a pipeline-based path tracer. This versatility makes our path tracing implementation highly adaptable, capable of delivering stunning and photorealistic images across a wide range of use-cases. Our solution sets a new standard in the field, combining speed, flexibility, and quality to meet the evolving demands of the graphics industry.

Micro-polygon rendering

With the advent of Nanite, a revolutionary development in the realm of graphical rendering, our team has embarked on an in-depth exploration into the sophisticated world of micro polygon rasterization techniques. Nanite enables the rendering of incredibly detailed virtual scenes with millions of polygons, something previously thought to be computationally prohibitive. Inspired by this breakthrough, our research is now focused on harnessing and further advancing these micro polygon rasterization methodologies