Используемая вами версия браузера не рекомендована для просмотра этого сайта.Установите последнюю версию браузера, перейдя по одной из следующих ссылок.
К сожалению, этот PDF-файл доступен только для загрузки
Digital Signal and Image Processing on Intel® ArchitectureExecutive SummaryGeneral purpose processors are used for digital signal and image processing (DSP) because of their flexibility, increasingly attractive performance/power ratios, and the generally lower cost to develop and support software. But there is increasing pressure to run more and more complex DSP algorithms on smaller and smaller platforms: Algorithms that required an HPC cluster a few years ago now need to run in a single rack mount server. And rack mount server applications now need to run on thin VPX or MicroTCA blades, for example.This paper examines the most important factors affecting DSP performance on Intel® Architecture-based processors and explains how the performance of an innovative radar post-processing algorithm was increased up to 33X by multithreading and using COTS multi-core hardware and performance libraries. It also discusses the new ability to tightly couple FPGAs to Intel processors and summarizes new instructions and microarchitecture that will roughly double the performance of many DSP algorithms on Intel processors that will release in 2010.Read the full Digital Signal and Image Processing on Intel® Architecture White Paper.
Maximize performance on a Windows*-based system using SMP and IntervalZero* RTX. (v.001, Oct. 2011)
Webinar: Reduce issues using Intel® SPDK and the included CriticalBlue* Prism* tool.
Tech leaders discuss the power of connecting billions of devices to extract greater big data insights.
Webinar details AltiVec* SIMD macros translator migration to Intel® processors. (v.1, May 2009)
Ishu Verma discusses Android* SDK and NDK toolsets as well as the benefits of using NDK. (v.001, Oct. 2011)