gpu1_2 - 4541.762 ( ) TA: (cicero at vplab.snu.ac.kr)...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
TA: 강성태 (cicero at vplab.snu.ac.kr) Grading Exam: 30% Term Project: 40% All information on the course will be available http://vplab.snu.ac.kr/lectures/09-2/graphics 4541.762 그래픽스 특강 ( 유저인터페이스 및 가시화 )
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Syllabus 1. Introduction to GPU 2. Rendering Pipeline 3. GPU Architecture Overview [GPU Architecture Overview (SIGGRAPH 2007).pdf, CUDA-book.pdf] Direct3D 9 and Shader Model 3 Part I [Direct3D 9 tutorial, DX9 Lecture Node 1, 2] 4. Direct3D 9 and Shader Model 3 Part II [Direct3D 9 tutorial, DX9 Lecture Node 3~5] 5. Direct3D 10 and Shader Model 4 [Introduction to Direct3D 10 (SIGGRAPH 07 Course).pdf] 6. Introduction to Parallel Programming [ Introduction to Parallel Programming Model (SIGGRAPH 2009).pdf] 7. High Level Language for GPU [High Level Languages for GPU.pdf] 8. Introduction to CUDA 1. Introduction to CUDA [Introduction to CUDA (NVIDIA).pdf] 2. CUDA Basics [CUDA Basics (NVIDIA).pdf] 9. Optimizing CUDA [Optimizing CUDA (NVIDIA).pdf] 10. App : Volume Rendering [VR Lecture notes] 11. App : Deformable Body Physics Simulation [Deformable Body Physics Simulation.pdf] 12. App : Marching Cubes [High-speed Marching Cubes using Histogram Pyramids.pdf] 13. App : Image Convolution [Image Convolution with CUDA (NVIDIA).pdf] 14. App : Edge Detection [Canny Edge Detection on NVIDIA CUDA (IEEE CVPRW 08).pdf] 15. App : Wavelet Transform [Discrete Wavelet Transform on GPU.pdf] 16. App : Image Denoising [Image Denoising (NVIDIA).pdf] 17. App : Image Registration [Accelerated Image Registration with CUDA.pdf] 18. App : Graph Cut [CudaCuts Fast Graph Cuts on the GPU (IEEE CVPRW 08).pdf]
Background image of page 2
Graphics Hardware Graphics Processing Unit (GPU) is a… Subsidiary hardware Independent to the main processing unit, i.e. CPU With massively multi-threaded many-core Hundreds of cores, thousands of concurrent threads Dedicated to 2D and 3D graphics Special purpose - low functionality, high performance H/W accelerated graphics operation
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Graphics Hardware CPUs are optimized for high performance on sequential code Branch prediction, out-of-order execution GPUs are optimized for highly data-parallel nature of graphics computation Plentiful SIMD instructions Extremely fast for SIMD operations Model for threading CPU - Coarse, heavyweight GPU - Fine, extremely lightweight
Background image of page 4
GPU in (relatively) Modern PCs
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
AGP bus 1x/2x/4x/8x 2133MB/s bandwidth with AGP 8x Asymmetric bandwidth CPU GPU download : almost 2GB/s on AGP 8x GPU CPU readback : <1/10 of download PCI Express 1x ~ 32x : just # of lanes Symmetric bandwidth for download/readback 1.1 : 0.25GB/s/lane for one direction (4GB/s for 16x) 2.0 : 0.5GB/s/lane 3.0 (in progress) : 0.8GB/s/lane (?) AGP/PCI Express Bus
Background image of page 6
GPUs are getting faster CPUs: annual growth 1.5 × decade growth : 60 × GPUs: annual growth > 2.0
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 49

gpu1_2 - 4541.762 ( ) TA: (cicero at vplab.snu.ac.kr)...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online