Software Rasterization of Large Triangles

Information

• Visibility: hidden
• Publication Type: Student Project
• Workgroup(s)/Project(s):
  • Rendering and Modeling
• Date: 2026
• Date (Start): 2025
• Date (End): 2026
• Matrikelnummer: 12433743
• First Supervisor: Markus Schütz
• Keywords: software rasterization

Abstract

This work explores a CUDA-based software rasterization approach that replaces the traditional hardware pipeline to improve flexibility and performance in specific rendering scenarios. While batch-based processing of triangles achieves high performance for dense scenes, it suffers from severe load imbalance when handling large primitives, leading to reduced parallel efficiency on modern GPUs. To address this, we implement a hierarchical tiled rasterization method using binning, which distributes the workload of large triangles across multiple thread blocks, significantly improving load balancing and overall performance.

Additional Files and Images

Weblinks

No further information available.

BibTeX

@studentproject{ioannis-2026-sro,
  title =      "Software Rasterization of Large Triangles",
  author =     "Liopas Ioannis",
  year =       "2026",
  abstract =   "This work explores a CUDA-based software rasterization
               approach that replaces the traditional hardware pipeline to
               improve flexibility and performance in specific rendering
               scenarios. While batch-based processing of triangles
               achieves high performance for dense scenes, it suffers from
               severe load imbalance when handling large primitives,
               leading to reduced parallel efficiency on modern GPUs. To
               address this, we implement a hierarchical tiled
               rasterization method using binning, which distributes the
               workload of large triangles across multiple thread blocks,
               significantly improving load balancing and overall
               performance.",
  keywords =   "software rasterization",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2026/ioannis-2026-sro/",
}