Design and evaluation of a multimedia computing architecture based on a 3D graphics pipeline

C. Y. Chung, R. Managuli, Yongmin Kim
{"title":"Design and evaluation of a multimedia computing architecture based on a 3D graphics pipeline","authors":"C. Y. Chung, R. Managuli, Yongmin Kim","doi":"10.1109/ASAP.2002.1030723","DOIUrl":null,"url":null,"abstract":"With the innovation and integration of media objects in multimedia applications, the importance of architectural support for different types of media objects, e.g., image, video and graphics, in one platform has significantly increased. While several approaches based on vector or VLIW (very long instruction word) architectures, e.g., Vector-IRAM and Imagine, have been pursued, they are not as effective as dedicated graphics pipelines for high-performance 3D graphics. We have explored a new programmable computing architecture based on a 3D graphics pipeline, which utilizes dedicated hardware resources in the 3D graphics pipeline for other types of multimedia computing. Adding programmable flexibility to a graphics pipeline for texture mapping has proven to be effective, e.g., pixel shader. However, due to the diversity of imaging and video processing applications, there are several challenges associated with converting a fixed graphics pipeline to a flexible multimedia computing engine. In this paper, we identify the additional architectural requirements, introduce the proposed architecture with extension details, and present the results of the performance evaluation. With cycle-accurate simulation of several benchmark functions, we have verified that the proposed architecture outperforms a modem powerful media processor in imaging and video processing by a factor of 1.3 to 7.5. The 3D graphics performance would not change much because the additional pipeline stages for the extension result in longer pipeline latency but similar throughout.","PeriodicalId":424082,"journal":{"name":"Proceedings IEEE International Conference on Application- Specific Systems, Architectures, and Processors","volume":"54 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings IEEE International Conference on Application- Specific Systems, Architectures, and Processors","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ASAP.2002.1030723","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

With the innovation and integration of media objects in multimedia applications, the importance of architectural support for different types of media objects, e.g., image, video and graphics, in one platform has significantly increased. While several approaches based on vector or VLIW (very long instruction word) architectures, e.g., Vector-IRAM and Imagine, have been pursued, they are not as effective as dedicated graphics pipelines for high-performance 3D graphics. We have explored a new programmable computing architecture based on a 3D graphics pipeline, which utilizes dedicated hardware resources in the 3D graphics pipeline for other types of multimedia computing. Adding programmable flexibility to a graphics pipeline for texture mapping has proven to be effective, e.g., pixel shader. However, due to the diversity of imaging and video processing applications, there are several challenges associated with converting a fixed graphics pipeline to a flexible multimedia computing engine. In this paper, we identify the additional architectural requirements, introduce the proposed architecture with extension details, and present the results of the performance evaluation. With cycle-accurate simulation of several benchmark functions, we have verified that the proposed architecture outperforms a modem powerful media processor in imaging and video processing by a factor of 1.3 to 7.5. The 3D graphics performance would not change much because the additional pipeline stages for the extension result in longer pipeline latency but similar throughout.
基于三维图形管道的多媒体计算体系结构设计与评价
随着多媒体应用中媒体对象的创新和集成,在一个平台中对不同类型的媒体对象(如图像、视频和图形)的架构支持的重要性显著增加。虽然有几种基于矢量或VLIW(非常长的指令字)架构的方法,例如vector - iram和Imagine,但它们并不像高性能3D图形的专用图形管道那样有效。我们探索了一种新的基于3D图形管道的可编程计算架构,它利用3D图形管道中的专用硬件资源进行其他类型的多媒体计算。在纹理映射的图形管道中添加可编程的灵活性已被证明是有效的,例如,像素着色器。然而,由于图像和视频处理应用程序的多样性,将固定的图形管道转换为灵活的多媒体计算引擎存在一些挑战。在本文中,我们确定了额外的体系结构需求,介绍了带有扩展细节的拟议体系结构,并给出了性能评估的结果。通过对几个基准功能的周期精确模拟,我们已经验证了所提出的架构在成像和视频处理方面比现代强大的媒体处理器性能高出1.3到7.5倍。3D图形性能不会有太大变化,因为扩展的额外管道阶段会导致更长的管道延迟,但整个过程是相似的。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
4.00
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信