嵌入式gpu中光栅化及深度预测试单元的分析与设计-analysis and design of raster and depth prediction test unit in embedded gpu.docxVIP

下载本文档

5
0
约4.59万字
约 59页
2018-07-28 发布于上海
举报
版权申诉

嵌入式gpu中光栅化及深度预测试单元的分析与设计-analysis and design of raster and depth prediction test unit in embedded gpu.docx

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

嵌入式gpu中光栅化及深度预测试单元的分析与设计-analysis and design of raster and depth prediction test unit in embedded gpu

ABSTRACTABSTRACTAsanimportantcomponentofmoderncomputersystem,graphicsprocessingsystemhasalwaysbornmassivecomputingtasks.Withthedevelopmentof3Dgraphicstechnolo-gy,graphicsprocessingsystemwithinembeddeddevicesfacesmoreandmorechallenges.TheembeddedGPUiscomposedofprogrammedShadersandRasterizationsystem,thelattermainlyinchargeofrenderingpixelsinsidebasicprimitivesandtheabilityofrender-ingpixelsisoneofthemostsignificantperformanceindicators.FortheembeddedGPU,thereisstillsomespaceforthescan-conversionalgorithmofrasterizationsystem.Lowren-deringtimeandpowerconsumptionstillneedtobesolvedtoguaranteetherenderingspeedandlowcost.Ontheotherhand,theembeddedGPUneedstoaccessexternalmemoryfre-quently,whichwastesmorebandwithandleadsmorepowerconsumption,anddepth-testoccupiesalmostahalf.Therefore,howtooptimizerenderingtime,powerconsumptionandmemorybandwithhasbecomeanimportantresearchdirection.Inthisdissertation,someresearchworksfocusingonaboveproblemsareproposed,includingscan-conversionalgorithmofrasterizationsystem,datapathofhardwaveoptimization,prototypesystemverificationandbandwithoptimizedarchitectureformemoryaccess,tosatisfytheneedsofrasterizationrenderinginembeddedGPU.First,anoptimizedalgorithmontraverseduring3Drenderingfortheembeddedde-vicesisproposed.Twosidesareimprovedbasedoncenterlineandtile-scanalgorithm,oneisforthetile-scan,pre-processingofsetupmoduleavoidsmanylogicaljudgmentsandtheotherisforthepixel-scan,edgedetectionforeachtileprocessesbeforepixel-scantoacceleratethespeedofscan-conversion.Second,weproposeahardwarearchitectureofrasterization.Floatingdesignmakeshighprecisionandpipelinestructuredesignacceleratestherenderingspeed.WeverifytheproposedrasterizationsystemintheformsofprototypesystembasedonFPGA,andobtaingoodspeedandareaperformance.Finally,anfastandhierarchicalEarlyZ-Testisproposedtorejectthepixelswhichareunnecessarytodrawassoonaspossiblefromthetilelevelandpixellevel.Redundantpre-pixeloperationsincludingZreads/writes,colorreads/writesandtexturereadsareavoidede?cientl