# ptina

**Repository Path**: archibate/ptina

## Basic Information

- **Project Name**: ptina
- **Description**: A path tracer based on Taichi
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-02-27
- **Last Updated**: 2023-06-26

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# PTina

A path tracer based on [Taichi](https://github.com/taichi-dev/taichi)

## How to run

For Linux & Mac OS X:
```py
python3 -m pip -U -r requirements.txt
export PYTHONPATH=`pwd`
python3 exams/interactive.py
```

For Windows 10:
```py
python -m pip -U -r requirements.txt
python -c "import os, sys, runpy; sys.path.append(os.getcwd()); runpy.run_path('exams/interactive.py')"
```

## Features

- Disney BSDF (with transmission)
- interactive camera control
- linear BVH for acceleration
- point, area, and environment lights
- textures, and a memory allocator for it
- material properties specified from textures
- multiple importance sampling (light & BSDF)
- Sobol quasi-random sequence generator
- metropilis light transport path tracing
- albedo & normal rendering as AOV
- Blender intergration as addon

## Performance

Unidirectional path tracer, 512x512 non-interactive, 32 samples for each pixel.
My hardware is GeForce 940MX, 2 GB VRAM, with CUDA 11.2 & driver 460.32.03.
The unit of numbers below is samples-per-second, the larger the better:

- 34 Tri, cornell with two boxes:

  * 5.17   OpenGL (local stack)
  * 6.54   OpenGL
  * 7.25   CUDA

- 978 Tri, cornell with a monkey:

  * 2.48   OpenGL (local stack)
  * 3.14   OpenGL
  * 2.88   CUDA
  * 2.22   CUDA (middle bvh)

## References

- https://github.com/wdas/brdf/blob/main/src/brdfs/disney.brdf
- http://www.pbr-book.org/3ed-2018/Reflection_Models/Microfacet_Models.html#eq:beckmann-lambda
- https://developer.nvidia.com/blog/thinking-parallel-part-iii-tree-construction-gpu/
- https://developer.nvidia.com/blog/thinking-parallel-part-ii-tree-traversal-gpu/
- https://web.maths.unsw.edu.au/~fkuo/sobol/