Particle System in WebGPU

Information

• Visibility: hidden
• Publication Type: Bachelor Thesis
• Workgroup(s)/Project(s): not specified
• Date: February 2023
• Date (Start): 2022
• Date (End): 2023
• Matrikelnummer: 00171283
• First Supervisor: Michael Wimmer

Abstract

"In this thesis, we investigate the performance of particle system simulation and rendering in WebGPU. A prototype that can produce various particle effects in the browser was implemented using the WebGPU API. The particle’s lifetimes and movements are simulated on the GPU using a compute shader, and rendered as textured quads. By using a compute shader, it is possible to simulate and render ten million particles at around 63 frames per second (on a GTX 1060). For rendering the particles, both instancing and vertex pulling were implemented, and a comparison between the two modes shows that on recent high end GPUs, using vertex pulling leads to significantly better performance than using instancing."

Additional Files and Images

Weblinks

No further information available.

BibTeX

@bachelorsthesis{PETER-2023-PSW,
  title =      "Particle System in WebGPU",
  author =     "Benedikt Peter",
  year =       "2023",
  abstract =   ""In this thesis, we investigate the performance of particle
               system simulation and rendering in WebGPU. A prototype that
               can produce various particle effects in the browser was
               implemented using the WebGPU API. The particle’s lifetimes
               and movements are simulated on the GPU using a compute
               shader, and rendered as textured quads. By using a compute
               shader, it is possible to simulate and render ten million
               particles at around 63 frames per second (on a GTX 1060).
               For rendering the particles, both instancing and vertex
               pulling were implemented, and a comparison between the two
               modes shows that on recent high end GPUs, using vertex
               pulling leads to significantly better performance than using
               instancing." ",
  month =      feb,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien ",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2023/PETER-2023-PSW/",
}