Advanced

Dynamic Ray Stream Traversal

Barringer, Rasmus LU and Akenine-Möller, Tomas LU (2014) 41th International Conference on Computer Graphics and Interactive Techniques, 2014 In ACM Transactions on Graphics 33(4). p.9-151
Abstract
While each new generation of processors gets larger caches and more compute power, external memory bandwidth capabilities increase at a much lower pace. Additionally, processors are equipped with wide vector units that require low instruction level divergence to be efficiently utilized. In order to exploit these trends in ray tracing, we present an alternative to traditional depth-first ray traversal that takes advantage of the available cache hierarchy, and provides high SIMD efficiency, while keeping memory bus traffic low. Our main contribution is an efficient algorithm for traversing large packets of rays against a bounding volume hierarchy in a way that groups coherent rays during traversal. In contrast to previous large packet... (More)
While each new generation of processors gets larger caches and more compute power, external memory bandwidth capabilities increase at a much lower pace. Additionally, processors are equipped with wide vector units that require low instruction level divergence to be efficiently utilized. In order to exploit these trends in ray tracing, we present an alternative to traditional depth-first ray traversal that takes advantage of the available cache hierarchy, and provides high SIMD efficiency, while keeping memory bus traffic low. Our main contribution is an efficient algorithm for traversing large packets of rays against a bounding volume hierarchy in a way that groups coherent rays during traversal. In contrast to previous large packet traversal methods, our algorithm allows for individual traversal order for each ray, which is essential for efficient ray tracing. Ray tracing algorithms is a mature research field in computer graphics, and despite this, our new technique increases traversal performance by 36-53%, and is applicable to most ray tracers. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
ACM Transactions on Graphics
volume
33
issue
4
pages
9 pages
publisher
ACM
conference name
41th International Conference on Computer Graphics and Interactive Techniques, 2014
external identifiers
  • wos:000340000100118
  • scopus:84905722732
ISSN
0730-0301
DOI
10.1145/2601097.2601222
language
English
LU publication?
yes
id
df76a230-24bb-4d56-a1aa-28f88917ab7a (old id 4588621)
date added to LUP
2014-08-19 11:55:37
date last changed
2017-11-05 04:11:04
@inproceedings{df76a230-24bb-4d56-a1aa-28f88917ab7a,
  abstract     = {While each new generation of processors gets larger caches and more compute power, external memory bandwidth capabilities increase at a much lower pace. Additionally, processors are equipped with wide vector units that require low instruction level divergence to be efficiently utilized. In order to exploit these trends in ray tracing, we present an alternative to traditional depth-first ray traversal that takes advantage of the available cache hierarchy, and provides high SIMD efficiency, while keeping memory bus traffic low. Our main contribution is an efficient algorithm for traversing large packets of rays against a bounding volume hierarchy in a way that groups coherent rays during traversal. In contrast to previous large packet traversal methods, our algorithm allows for individual traversal order for each ray, which is essential for efficient ray tracing. Ray tracing algorithms is a mature research field in computer graphics, and despite this, our new technique increases traversal performance by 36-53%, and is applicable to most ray tracers.},
  author       = {Barringer, Rasmus and Akenine-Möller, Tomas},
  booktitle    = {ACM Transactions on Graphics},
  issn         = {0730-0301},
  language     = {eng},
  number       = {4},
  pages        = {9--151},
  publisher    = {ACM},
  title        = {Dynamic Ray Stream Traversal},
  url          = {http://dx.doi.org/10.1145/2601097.2601222},
  volume       = {33},
  year         = {2014},
}