Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration
碩士 === 國立交通大學 === 多媒體工程研究所 === 100 === In this work, an energy-efficient programmable vertex processor with ray-tracing acceleration is proposed. The proposed vertex processor architecture has four features: 1) four-issue floating-point VLIW datapath; 2) multilevel pre-TnL and post-TnL vertex cache...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2012
|
Online Access: | http://ndltd.ncl.edu.tw/handle/79716181762820169949 |
id |
ndltd-TW-100NCTU5641034 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-100NCTU56410342016-03-28T04:20:35Z http://ndltd.ncl.edu.tw/handle/79716181762820169949 Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration 可支援光線追蹤技術之可程式化高能源效率頂點處理器之設計與實現 Liu, Ruei-Jyun 劉睿峻 碩士 國立交通大學 多媒體工程研究所 100 In this work, an energy-efficient programmable vertex processor with ray-tracing acceleration is proposed. The proposed vertex processor architecture has four features: 1) four-issue floating-point VLIW datapath; 2) multilevel pre-TnL and post-TnL vertex cache functions; 3) Whitted-style ray-tracing supported; 4) hardware sharing between the intersection and back-face culling processing. From the chip implementation result in TSMC 90nm CMOS process technology, the operating frequency can be up to 100MHz. The post-layout simulation indicates that the vertex fill rate with and without lighting can achieve 4.5Mvertices/s and 100Mvertices/s, respectively. The resulting power efficiency is 268.33Kvertices/mJ. Van, Lan-Da 范倫達 2012 學位論文 ; thesis 43 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立交通大學 === 多媒體工程研究所 === 100 === In this work, an energy-efficient programmable vertex processor with ray-tracing acceleration is proposed. The proposed vertex processor architecture has four features: 1) four-issue floating-point VLIW datapath; 2) multilevel pre-TnL and post-TnL vertex cache functions; 3) Whitted-style ray-tracing supported; 4) hardware sharing between the intersection and back-face culling processing. From the chip implementation result in TSMC 90nm CMOS process technology, the operating frequency can be up to 100MHz. The post-layout simulation indicates that the vertex fill rate with and without lighting can achieve 4.5Mvertices/s and 100Mvertices/s, respectively. The resulting power efficiency is 268.33Kvertices/mJ.
|
author2 |
Van, Lan-Da |
author_facet |
Van, Lan-Da Liu, Ruei-Jyun 劉睿峻 |
author |
Liu, Ruei-Jyun 劉睿峻 |
spellingShingle |
Liu, Ruei-Jyun 劉睿峻 Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
author_sort |
Liu, Ruei-Jyun |
title |
Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
title_short |
Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
title_full |
Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
title_fullStr |
Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
title_full_unstemmed |
Design and Implementation of an Energy-Efficient Programmable Vertex Processor with Ray-Tracing Acceleration |
title_sort |
design and implementation of an energy-efficient programmable vertex processor with ray-tracing acceleration |
publishDate |
2012 |
url |
http://ndltd.ncl.edu.tw/handle/79716181762820169949 |
work_keys_str_mv |
AT liurueijyun designandimplementationofanenergyefficientprogrammablevertexprocessorwithraytracingacceleration AT liúruìjùn designandimplementationofanenergyefficientprogrammablevertexprocessorwithraytracingacceleration AT liurueijyun kězhīyuánguāngxiànzhuīzōngjìshùzhīkěchéngshìhuàgāonéngyuánxiàolǜdǐngdiǎnchùlǐqìzhīshèjìyǔshíxiàn AT liúruìjùn kězhīyuánguāngxiànzhuīzōngjìshùzhīkěchéngshìhuàgāonéngyuánxiàolǜdǐngdiǎnchùlǐqìzhīshèjìyǔshíxiàn |
_version_ |
1718212847294480384 |