TraverseResearchWork with us
Menu

Traverse Research

Micro-polygon rendering

Detailed geometry, visibility buffers, mesh shading and asset preparation investigated as one rendering pipeline.

Micro-polygon research from Traverse’s Breda framework.
Micro-polygon research from Traverse’s Breda framework.

Detail that reaches the frame selectively

High-detail geometry makes visibility, representation and data movement first-class rendering problems. The amount of source geometry is not the same as the amount of geometry that should be processed for a particular image.

Breda’s research combines micro-polygon techniques, mesh shading, visibility-buffer approaches and an in-house asset pipeline. The point of integration is to investigate the whole path from prepared assets to the work executed on the GPU.

Demonstrations and workload design

Our One Trillion Triangles demonstration explores geometry at scale in Breda. It connects the representation of a large scene to the visibility and streaming work needed to render a particular view.

Geometry demonstration · Inside the Breda framework.

A few useful answers.

Is this only a runtime rendering problem?

No. How geometry is prepared, represented and streamed is part of the pipeline.

Does the demonstration title imply a measured frame rate?

No. The title describes its geometry-scale theme; hardware-specific performance needs its own measured conditions.

Can this inform a mobile workload?

Yes. Geometry processing, visibility and streaming need to be designed around the target device’s memory, bandwidth and frame-time budgets. We investigate the approach in the context of the renderer and content pipeline.

Traverse Research + Traverse Games

Let’s turn a good question into working software.

Start a conversation