Ray Tracing on Programmable Graphics Hardware

1
Ray Tracing on Programmable Graphics Hardware
SIGGRAPH 2002

• Timothy J. Purcell Ian Buck
• William R. Mark Pat Hanrahan

Sungbae Kim Master Student in CS
2
Contributions

• Abstract streaming graphics processor
• Implement ray tracing using GPU
• Help guide the design of future architectures

3
Why ray tracing?

• Ray tracing for the movie Cars

4
Ray tracing pros and cons

• Good quality
• Simple algorithm
• Expensive cost
• ? How to accelerate ray tracing

5
Basic Ray tracing algorithm

• Shoot eye rays
• Find the first intersection
• Calculates the illumination
• Reflection? Calls itself
• Transmission? Calls itself
• Combines the resulting illumination

6
Basic Ray tracing

?

Note that each pixel is independent
7
Find intersection
8
Uniform Grid
9
How to accelerate

• Many algorithms
• Use parallel systems
• Massively parallel shared memory supercomputer
  -ray
• Cluster of commodity PCs RTRT
• Special hardware for ray tracing
• Streaming Programming Model this papers
  approach

10
Streaming Programming Model

• A stream is a set of data records
• Kernels operate on each element of an input
  stream independently
• Kernels can read from global memory
• Streams connect kernels

11
Streaming Ray Tracer
Screen Pixels
Eye Ray Generator
Camera
Rays
Traverser
Grid
(Ray, Voxel)
Intersector
Triangles
Hits
Shader
Materials
Pixel color updates
12
Programmable Graphics Processor Abstraction

• We can use GPU as Streaming Processor
• But limited streaming processor

13
GPU Ray Tracer

• Limited ? How to map
• Global Memory
• Kernel to Kernel connection
• No loop, No branch

14
Texture Memory Mapping

• We can use texture as global memory
• Store non-RGB values in textures

For example, The Second triangle of the second
grind
Also Rays, Materials
15
Kernel to Kernel

• Copy output fragments into Texture

16
No loop No Branch

• Multipass operate again until every is done
• Do
• bodyfp()
• while ( donefp() )
• State z-value, stencil bit
• For performance early z-culling, stencil mask

17
Simulation Result

• SIMD vs. MIMD

18
GPU vs. CPU
Radeon 9700 Pro CPU
Ray-triangle intersections/s 100M 20M in P3 800 MHz
Ray/s 300K to 4.0M 800K to 7.1M 1.8M to 2.3M (With simple shading) in P4 2.5 GHz

• Comparable to CPU

19
GPU vs. CPU

• Graphics processor evolves faster

20
Demo
Shadow rays 10.5 fps
Eye rays, Shadow rays, Reflection rays 1.0 fps
Shadow rays 1.8 fps
21
AppendixGeneral Purpose GPU(GPGPU)

• GPU, Shader, GLSL, HLSL, Cg, OpenGL, Direct3D.
• BrookGPU (Stanford)
• CUDA (NVIDIA)

22
AppendixCell Processor

• IBM, SONY, TOSHIBA
• Interactive ray tracer iRT
• More faster than GPU