Multi Bounding Volume Hierarchies for Ray Tracing Pipelines

Timo Viitanen; Matias Koskela; Pekka Jääskeläinen; Jarmo Takala

doi:10.1145/3005358.3005384

Multi Bounding Volume Hierarchies for Ray Tracing Pipelines

Timo Viitanen
, Matias Koskela
, Pekka Jääskeläinen
, Jarmo Takala

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

8 Citations (Scopus)

568 Downloads (Pure)

Abstract

High-performance ray tracing on CPU is now largely based on Multi Bounding Volume Hierarchy (MBVH) trees. We apply MBVH to a fixed-function ray tracing accelerator architecture. According to cycle-level simulations and power analysis, MBVH reduces energy per frame by an average of 24% and improves performance per area by 19% in scenes with incoherent rays, due to its compact memory layout which reduces DRAM traffic. With primary rays, energy efficiency improves by 15% and performance per area by 20%.

Original language	English
Title of host publication	SIGGRAPH ASIA 2016 Technical Briefs
Publisher	ACM
ISBN (Print)	978-1-4503-4541-5
DOIs	https://doi.org/10.1145/3005358.3005384
Publication status	Published - 2016
Publication type	A4 Article in conference proceedings
Event	ACM SIGGRAPH Asia - Duration: 1 Jan 1900 → …

Conference

Conference	ACM SIGGRAPH Asia
Period	1/01/00 → …

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

ray tracing, ray tracing hardware

Publication forum classification

Publication forum level 1

Access to Document

10.1145/3005358.3005384

Camera ready manuscript
© ACM 2016. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Proceeding SA '16 SIGGRAPH ASIA 2016 Technical Briefs , http://dx.doi.org/10.1145/3005358.3005384
Accepted author manuscript, 5.51 MB

http://urn.fi/URN:NBN:fi:tty-201612164871

Cite this

@inproceedings{1184ca4ae5d14b3d8125a616d477541a,

title = "Multi Bounding Volume Hierarchies for Ray Tracing Pipelines",

abstract = "High-performance ray tracing on CPU is now largely based on Multi Bounding Volume Hierarchy (MBVH) trees. We apply MBVH to a fixed-function ray tracing accelerator architecture. According to cycle-level simulations and power analysis, MBVH reduces energy per frame by an average of 24\% and improves performance per area by 19\% in scenes with incoherent rays, due to its compact memory layout which reduces DRAM traffic. With primary rays, energy efficiency improves by 15\% and performance per area by 20\%.",

keywords = "ray tracing, ray tracing hardware",

author = "Timo Viitanen and Matias Koskela and Pekka J{\"a}{\"a}skel{\"a}inen and Jarmo Takala",

year = "2016",

doi = "10.1145/3005358.3005384",

language = "English",

isbn = "978-1-4503-4541-5",

booktitle = "SIGGRAPH ASIA 2016 Technical Briefs",

publisher = "ACM",

address = "United States",

note = "ACM SIGGRAPH Asia ; Conference date: 01-01-1900",

}

TY - GEN

T1 - Multi Bounding Volume Hierarchies for Ray Tracing Pipelines

AU - Viitanen, Timo

AU - Koskela, Matias

AU - Jääskeläinen, Pekka

AU - Takala, Jarmo

PY - 2016

Y1 - 2016

N2 - High-performance ray tracing on CPU is now largely based on Multi Bounding Volume Hierarchy (MBVH) trees. We apply MBVH to a fixed-function ray tracing accelerator architecture. According to cycle-level simulations and power analysis, MBVH reduces energy per frame by an average of 24% and improves performance per area by 19% in scenes with incoherent rays, due to its compact memory layout which reduces DRAM traffic. With primary rays, energy efficiency improves by 15% and performance per area by 20%.

AB - High-performance ray tracing on CPU is now largely based on Multi Bounding Volume Hierarchy (MBVH) trees. We apply MBVH to a fixed-function ray tracing accelerator architecture. According to cycle-level simulations and power analysis, MBVH reduces energy per frame by an average of 24% and improves performance per area by 19% in scenes with incoherent rays, due to its compact memory layout which reduces DRAM traffic. With primary rays, energy efficiency improves by 15% and performance per area by 20%.

KW - ray tracing, ray tracing hardware

U2 - 10.1145/3005358.3005384

DO - 10.1145/3005358.3005384

M3 - Conference contribution

SN - 978-1-4503-4541-5

BT - SIGGRAPH ASIA 2016 Technical Briefs

PB - ACM

T2 - ACM SIGGRAPH Asia

Y2 - 1 January 1900

ER -

Multi Bounding Volume Hierarchies for Ray Tracing Pipelines

Abstract

Conference

UN SDGs

Keywords

Publication forum classification

Access to Document

Fingerprint

Cite this