TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

PLOCTree: A Fast, High-Quality Hardware BVH Builder

Tutkimustuotosvertaisarvioitu

Standard

PLOCTree: A Fast, High-Quality Hardware BVH Builder. / Viitanen, Timo; Koskela, Matias; Jääskeläinen, Pekka; Tervo, Aleksi; Takala, Jarmo.

julkaisussa: Proceedings of the ACM on Computer Graphics and Interactive Techniques, Vuosikerta 1, Nro 2, 12.08.2018.

Tutkimustuotosvertaisarvioitu

Harvard

Viitanen, T, Koskela, M, Jääskeläinen, P, Tervo, A & Takala, J 2018, 'PLOCTree: A Fast, High-Quality Hardware BVH Builder', Proceedings of the ACM on Computer Graphics and Interactive Techniques, Vuosikerta. 1, Nro 2. https://doi.org/10.1145/3233309

APA

Viitanen, T., Koskela, M., Jääskeläinen, P., Tervo, A., & Takala, J. (2018). PLOCTree: A Fast, High-Quality Hardware BVH Builder. Proceedings of the ACM on Computer Graphics and Interactive Techniques, 1(2). https://doi.org/10.1145/3233309

Vancouver

Viitanen T, Koskela M, Jääskeläinen P, Tervo A, Takala J. PLOCTree: A Fast, High-Quality Hardware BVH Builder. Proceedings of the ACM on Computer Graphics and Interactive Techniques. 2018 elo 12;1(2). https://doi.org/10.1145/3233309

Author

Viitanen, Timo ; Koskela, Matias ; Jääskeläinen, Pekka ; Tervo, Aleksi ; Takala, Jarmo. / PLOCTree: A Fast, High-Quality Hardware BVH Builder. Julkaisussa: Proceedings of the ACM on Computer Graphics and Interactive Techniques. 2018 ; Vuosikerta 1, Nro 2.

Bibtex - Lataa

@article{1712e3155b99448bafc00cd207517ac9,
title = "PLOCTree: A Fast, High-Quality Hardware BVH Builder",
abstract = "In the near future, GPUs are expected to have hardware support for real-time ray tracing in order to, e.g., help render complex lighting effects in video games and enable photorealistic augmented reality. One challenge in real-time ray tracing is dynamic scene support, that is, rebuilding or updating the spatial data structures used to accelerate rendering whenever the scene geometry changes. This paper proposes PLOCTree, an accelerator for tree construction based on the Parallel Locally-Ordered Clustering (PLOC) algorithm. Tree construction is highly memory-intensive, thus for the hardware implementation, the algorithm is rewritten into a bandwidtheconomical form which converts most of the external memory traffic of the original software-based GPU implementation into streaming on-chip data traffic. As a result, the proposed unit is 3.9 times faster and uses 7.7 times less memory bandwidth than the GPU implementation. Compared to state-of-the-art hardware builders, PLOCTree gives a superior performance-quality tradeoff: it is nearly as fast as a state-of-the-art low-quality linear builder, while producing trees of similar Surface Area Heuristic (SAH) cost as a comparatively expensive binned SAH sweep builder.",
author = "Timo Viitanen and Matias Koskela and Pekka J{\"a}{\"a}skel{\"a}inen and Aleksi Tervo and Jarmo Takala",
year = "2018",
month = "8",
day = "12",
doi = "10.1145/3233309",
language = "English",
volume = "1",
journal = "Proceedings of the ACM on Computer Graphics and Interactive Techniques",
issn = "2577-6193",
publisher = "ACM",
number = "2",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - PLOCTree: A Fast, High-Quality Hardware BVH Builder

AU - Viitanen, Timo

AU - Koskela, Matias

AU - Jääskeläinen, Pekka

AU - Tervo, Aleksi

AU - Takala, Jarmo

PY - 2018/8/12

Y1 - 2018/8/12

N2 - In the near future, GPUs are expected to have hardware support for real-time ray tracing in order to, e.g., help render complex lighting effects in video games and enable photorealistic augmented reality. One challenge in real-time ray tracing is dynamic scene support, that is, rebuilding or updating the spatial data structures used to accelerate rendering whenever the scene geometry changes. This paper proposes PLOCTree, an accelerator for tree construction based on the Parallel Locally-Ordered Clustering (PLOC) algorithm. Tree construction is highly memory-intensive, thus for the hardware implementation, the algorithm is rewritten into a bandwidtheconomical form which converts most of the external memory traffic of the original software-based GPU implementation into streaming on-chip data traffic. As a result, the proposed unit is 3.9 times faster and uses 7.7 times less memory bandwidth than the GPU implementation. Compared to state-of-the-art hardware builders, PLOCTree gives a superior performance-quality tradeoff: it is nearly as fast as a state-of-the-art low-quality linear builder, while producing trees of similar Surface Area Heuristic (SAH) cost as a comparatively expensive binned SAH sweep builder.

AB - In the near future, GPUs are expected to have hardware support for real-time ray tracing in order to, e.g., help render complex lighting effects in video games and enable photorealistic augmented reality. One challenge in real-time ray tracing is dynamic scene support, that is, rebuilding or updating the spatial data structures used to accelerate rendering whenever the scene geometry changes. This paper proposes PLOCTree, an accelerator for tree construction based on the Parallel Locally-Ordered Clustering (PLOC) algorithm. Tree construction is highly memory-intensive, thus for the hardware implementation, the algorithm is rewritten into a bandwidtheconomical form which converts most of the external memory traffic of the original software-based GPU implementation into streaming on-chip data traffic. As a result, the proposed unit is 3.9 times faster and uses 7.7 times less memory bandwidth than the GPU implementation. Compared to state-of-the-art hardware builders, PLOCTree gives a superior performance-quality tradeoff: it is nearly as fast as a state-of-the-art low-quality linear builder, while producing trees of similar Surface Area Heuristic (SAH) cost as a comparatively expensive binned SAH sweep builder.

U2 - 10.1145/3233309

DO - 10.1145/3233309

M3 - Article

VL - 1

JO - Proceedings of the ACM on Computer Graphics and Interactive Techniques

JF - Proceedings of the ACM on Computer Graphics and Interactive Techniques

SN - 2577-6193

IS - 2

ER -