Accelerated Single Ray Tracing for Wide Vector Units

Simple item page
dc.contributor.authorFuetterling, Valentinen_US
dc.contributor.authorLojewski, Carstenen_US
dc.contributor.authorPfreundt, Franz-Josefen_US
dc.contributor.authorHamann, Bernden_US
dc.contributor.authorEbert, Achimen_US
dc.contributor.editorVlastimil Havran and Karthik Vaiyanathanen_US
dc.date.accessioned2017-12-06T19:47:30Z
dc.date.available2017-12-06T19:47:30Z
dc.date.issued2017
dc.description.abstractUtilizing the vector units of current processors for ray tracing single rays through Bounding Volume Hierarchies has been accomplished by increasing the branching factor of the acceleration structure to match the vector width. A high branching factor allows vectorized bounding box tests but requires a complex control flow for the calculation of a front-to-back traversal order. We propose a novel algorithm for single rays entirely based on vector operations that performs a complete traversal iteration in constant time, ideally suited for current and future micro architectures featuring wide vector units. In addition we use our single ray technique as a building block to construct a fast packet traversal for coherent rays. We validate our algorithms with implementations utilizing the AVX2 and AVX-512 instruction sets and demonstrate significant performance gains over state-of-the-art solutions.en_US
dc.description.sectionheadersRay Traversal and Intersection
dc.description.seriesinformationEurographics/ ACM SIGGRAPH Symposium on High Performance Graphics
dc.identifier.doi10.1145/3105762.3105785
dc.identifier.isbn978-1-4503-5101-0
dc.identifier.issn2079-8679
dc.identifier.urihttps://doi.org/10.1145/3105762.3105785
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1145/3105762-3105785
dc.publisherACMen_US
dc.subjectComputing methodologies
dc.subjectRay tracing
dc.subjectray tracing
dc.subjectspatial data structure
dc.subjectbounding volume hierarchy
dc.subjectvectorized data processing
dc.titleAccelerated Single Ray Tracing for Wide Vector Unitsen_US
Files
Collections
High-Performance Graphics 2017