{"title":"Experiments Around Sparc Leon-2 For Mpeg Encoding","authors":"P. Guironnet De Massas, P. Amblard","doi":"10.1109/MIXDES.2006.1706585","DOIUrl":null,"url":null,"abstract":"In this paper, we present a modified implementation of the Sparc R V8 architecture with multimedia oriented instructions. The purpose of the global project is to improve video compression performances without designing a dedicated co-processor or using an expensive system on chip. In this project we investigate the complexity of modifying a standard processor. We implemented 12 new instructions and use some of them to optimize an algorithm of `Sum of Absolute Difference'. The result was a performance increase by 70% in the execution of this algorithm, allowing us to expect 30% speedup in the execution of a MPEG video compression. The area increase of the integer unit is about 18% and the clock frequency is not significantly modified in a LEON-2 implementing 5 from 12 of the new instructions","PeriodicalId":318768,"journal":{"name":"Proceedings of the International Conference Mixed Design of Integrated Circuits and System, 2006. MIXDES 2006.","volume":"33 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2006-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the International Conference Mixed Design of Integrated Circuits and System, 2006. MIXDES 2006.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MIXDES.2006.1706585","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
In this paper, we present a modified implementation of the Sparc R V8 architecture with multimedia oriented instructions. The purpose of the global project is to improve video compression performances without designing a dedicated co-processor or using an expensive system on chip. In this project we investigate the complexity of modifying a standard processor. We implemented 12 new instructions and use some of them to optimize an algorithm of `Sum of Absolute Difference'. The result was a performance increase by 70% in the execution of this algorithm, allowing us to expect 30% speedup in the execution of a MPEG video compression. The area increase of the integer unit is about 18% and the clock frequency is not significantly modified in a LEON-2 implementing 5 from 12 of the new instructions
在本文中,我们提出了一个基于多媒体指令的Sparc R V8架构的改进实现。全球项目的目的是在不设计专用协处理器或使用昂贵的芯片系统的情况下提高视频压缩性能。在这个项目中,我们研究修改标准处理器的复杂性。我们实现了12条新指令,并使用其中一些指令来优化“绝对差和”算法。结果是,在执行该算法时,性能提高了70%,使我们能够期望在执行MPEG视频压缩时加速30%。整数单元的面积增加了约18%,并且在从12条新指令中实现5的LEON-2中,时钟频率没有显着改变