Chinaunix首页 | 论坛 | 博客
  • 博客访问: 41339
  • 博文数量: 47
  • 博客积分: 1000
  • 博客等级: 少尉
  • 技术积分: 490
  • 用 户 组: 普通用户
  • 注册时间: 2009-12-02 21:19
文章分类
文章存档

2010年(1)

2009年(46)

我的朋友
最近访客

分类:

2009-12-04 00:53:37

GPU: A closer look
 
Geometry primitive is represented by groups of   vertex, application provides (x,y,z) to VG stage   
         
VP stage chage 3D position into 2D position
 
PG group vertex together to into order stream to form primitives              
 
PP application dependent, generate new order stream
 
FG sample each primitive, output fragment record
about the image position on the surface, the distance to the virtual camera
 
FP give surfaces realistic appearances
 
PO apply fragment contribution to output image pixel value
 
 
 
shader functions define the behavior of application programmable stage (vp, pp, fp). They are compiled into bytecode offline and tranlated into binary code specific to a GPU in runtime. It does not execute parallel but only serially process inputs ( I don't understand here...)
 
Graphics processing present large amount of data and task level parallelism, but the PO is a serial point where there are a lot of find grain dependence.
The bandwidth and computational demand vary significant for different load, so it is necessary to map the processing and storage resource to pipeline stage dynamically.
 
 
 
阅读(447) | 评论(0) | 转发(0) |
给主人留下些什么吧!~~