Dissertation > Excellent graduate degree dissertation topics show

Coarse-Z Filter:A Low-bandwidth Algorithm and Its Hardening for Hidden Surface Removal

Author: GaoKe
Tutor: ShiJiaoYing
School: Zhejiang University
Course: Computer Applications
Keywords: The graphics pipeline Bandwidth bottlenecks Blanking algorithm Coarse - Z Filter Bandwidth model
CLC: TP391.41
Type: Master's thesis
Year: 2006
Downloads: 58
Quote: 0
Read: Download Dissertation

Abstract


Graphics hardware in fixed graphics pipeline mainly consider three questions: First, the storage space, mainly used for storing the draw objects texture data, in-depth data and color data; computing power, including the floating of the geometry processing stage point operations and the rasterization stage of the vertex operator; bandwidth capabilities, including reading the texture cache, read and write the depth buffer and frame buffer read and write the required bandwidth. With the rapid decline in memory prices and the rapid increase of GPU computing power, storage space problems and computing power has been effectively alleviated. Graphics hardware capacity of the bandwidth requirements of computing power, under normal circumstances, the higher computing power will inevitably bring higher bandwidth requirements. GPU development much faster than the pace of development of the graphics hardware interface, the current mainstream graphics hardware, relative to the computing power of the graphics hardware, bandwidth capacity is insufficient, become a bottleneck. This paper describes the additional hardware unit Coarse-ZFilter designed to optimize the bandwidth of the graphics pipeline, to establish a mathematical model of the using Coarse-Z Filter the graphics pipeline bandwidth, and in accordance with the mathematical model, using standard industry test software, test Coarse- Z Filter efficiency, quantitative analysis of the main factors affect the bandwidth of the graphics pipeline, in order to assess the optimal parameters of the Coarse-Z Filter. Coarse-Z Filter the core idea is to use a low-resolution filter cache unit organization Tile, Tile-level depth testing before the pixel-level depth testing to reduce the number of read Z-Buffer, thus effectively reducing the graphics pipeline bandwidth. Filter cache the scan conversion Tile the maximum depth of the pixel value is updated, the filter cache update strategy, Coarse-Z Filter has two modes, that is, a single cache mode and double-buffered mode. For the quantitative analysis of Coarse-Z Filter optimize the contribution of the graphics pipeline bandwidth, Coarse-Z Filter bandwidth model. Bandwidth model, the optimization of the graphics pipeline bandwidth mainly depends on two factors, namely of Coarse-Z Filter the efficiency and Coarse-Z Filter bandwidth consumption. These two factors vary depending on the the Coarse-Z Filter the use of two parameters, the Tile size n bit format k and filter the cache. Used herein, the standard industry test software, the type (n, k) were tested to determine the optimal parameters. The test results show that, Coarse-Z Filter on the graphics pipeline bandwidth optimization effect is most obvious in the case of tile size 4x4, filter cache 16bits. At this time, the left of the use can be reduced by 30% Coarse-Z Filter

Related Dissertations

  1. Design and Realize the IP Core of 3D Embedded Graphics Processing Unit Based on FPGA,TP391.41
  2. Design of Embedded Graphics Processing Unit Based on ARM and FPGA,TP391.41
  3. Research about 3D Graphics Engine and Its Application Development on Mobile Brew Platform of CDMA System,TP391.41
  4. Research and Implementation of the Embedded Graphic Processor Unit,TP391.41
  5. Large-Scale Network Traffic Bottleneck Analysis,TP393.06
  6. The Design and Implementation of the Joint Simulation Training System,TP391.9
  7. 3D graphics engine key technology research,TP391.41
  8. Non-Photorealistic Rendering Techniques Based on Programmable Graphics Hardware,TP391.41
  9. The Geometric Modeling System for 3-D Packaging CAD,TP391.92
  10. NC simulation modeling and technology research,TG659
  11. Real-time Rendering on GPU,TP391.41
  12. Design of Graphics Processing Unit for Embedded Systems,TP368.1
  13. The Design & Realization of Mirror Load & Accelerated Access System of Jilin Provice Government Potorl,TP393.092
  14. Low- voltage , low -power CMOS pipelined ADC and Implementation,TN792
  15. Research on Basic Algorithms of Digital Image Processing and Implementation with FPGA,TP391.41
  16. Research on Facial Feature Extraction and Matching Algorithms for Image Retrieval,TP391.41
  17. Research of High Speed Image Pre-processing System Based on FPGA,TP391.41
  18. Research on Algorithms of 2D Face Template Protection,TP391.41
  19. The Research and Implemention of Image Retrieval Based on User Interested Feature,TP391.41
  20. Research of Image Mosaic Technology,TP391.41
  21. Research and Implementation of Exact String Matchiing Algorithms,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile