John Redford, Bret Bersack, M. Monk, Fred Huettig, D. Fitzgerald
{"title":"A vector DSP for imaging","authors":"John Redford, Bret Bersack, M. Monk, Fred Huettig, D. Fitzgerald","doi":"10.1109/CICC.2002.1012788","DOIUrl":null,"url":null,"abstract":"The CW4011 is an SoC containing a DSP with a novel vector architecture. It exploits the parallelism and narrow data typical of image processing to gain high performance at a low cost and power. It contains eight 32-bit datapaths all working off of a single instruction. It can do 16 16-bit MACs/cycle, and also four 32-bit memory accesses per cycle to 128 KB of on-chip memory. It also contains a serial datapath for handling low-performance code and OS functions. The chip also includes memory, video, and IO interfaces on an industry-standard bus. it is built in 0.18 /spl mu/m CMOS technology, is 7.8 /spl times/ 6.8 mm, runs at 200 MHz (worst-case) and draws less than 500 mW. This gives the best cost-performance of any processor on the market for imaging applications.","PeriodicalId":209025,"journal":{"name":"Proceedings of the IEEE 2002 Custom Integrated Circuits Conference (Cat. No.02CH37285)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the IEEE 2002 Custom Integrated Circuits Conference (Cat. No.02CH37285)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CICC.2002.1012788","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
The CW4011 is an SoC containing a DSP with a novel vector architecture. It exploits the parallelism and narrow data typical of image processing to gain high performance at a low cost and power. It contains eight 32-bit datapaths all working off of a single instruction. It can do 16 16-bit MACs/cycle, and also four 32-bit memory accesses per cycle to 128 KB of on-chip memory. It also contains a serial datapath for handling low-performance code and OS functions. The chip also includes memory, video, and IO interfaces on an industry-standard bus. it is built in 0.18 /spl mu/m CMOS technology, is 7.8 /spl times/ 6.8 mm, runs at 200 MHz (worst-case) and draws less than 500 mW. This gives the best cost-performance of any processor on the market for imaging applications.