Skip to content

Engineering thesis project tackling path tracing in a remote, multi-GPU environment.

Notifications You must be signed in to change notification settings

3DevApps/multi-gpu-path-tracer

Repository files navigation

Multi-GPU Path Tracer

Setup

Run before building dependencies from source:

Ares

source scripts/ares_setup.sh

Ubuntu

source scripts/ubuntu_setup.sh

Building dependencies

scripts/install_deps.sh downloads specified releases, builds them and installs them in ~/libs.

./scripts/install_deps.sh

Compiling and Running

To compile and run the project, follow the steps below:

  1. Create a build directory
mkdir build
  1. Build the project
cmake --build build

Note

If you get Cmake Error: could not load cache error, run cmake -S . -B build and re-run the build command.

  1. Run the project
build/cuda_project

Visualizing the output

The output file is saved in the PPM format. To visualize it, you can utilize the GNOME Image Viewer.

eog build/out.ppm

Using custom .obj and .mtl files

To use custom .obj and .mtl files, specify the paths to these files in the main.cpp file.

Currently, the program only supports the following materials with the corresponding properties:

  • lambertian
    • Ka
  • metal
    • Ka
    • Ns
  • dielectric
    • Ni
  • diffuse_light
    • Kd

To use these materials, ensure that the material's name in the .mtl file matches with one from the list above.

About

Engineering thesis project tackling path tracing in a remote, multi-GPU environment.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published