The Reference Path Tracer

Boksansky, Jakub; Marrs, Adam

doi:10.1007/978-1-4842-7185-8_14

Search citation statements

Order By: Relevance

Paper Sections

Select...

Background and Related Work1

Rays Paths Hits and Ownership Masks1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2022

2023

Publication Types

Select...

Article1

Other1

Relationship

Self Cite0

Independent2

Authors

Journals

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We assume familiarity with the basics of modern ray and path tracing; for reference, we point the reader to Alarcon's explanation of the RTX pipeline [Ala20], and Boksanksy and Marrs' Reference Path Tracer [BM21]. We also assume familiarity with the concept of wavefront path tracing (see, e.g., [LKA13]).…”

Section: Background and Related Workmentioning

confidence: 99%

Data Parallel Multi‐GPU Path Tracing using Ray Queue Cycling

Wald¹,

Jaroš

Zellmann

2023

Computer Graphics Forum

View full text Add to dashboard Cite

We propose a novel approach to data‐parallel path tracing on single‐node/multi‐GPU hardware that builds on ray forwarding, but which aims—above all else—at generality and practicability. We do this by avoiding any attempts at reducing the number of traces or forward operations performed, and instead focus on always using all GPUs' aggregate compute and bandwidth to effectively trace each ray on every GPU. We show that—counter‐intuitively—this is both feasible and desirable; and that when run on typical data‐center/cloud hardware, the resulting framework not only achieves good performance and scalability, but also comes with significantly fewer limitations, assumptions, or preprocessing requirements than existing techniques.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Data Parallel Multi‐GPU Path Tracing using Ray Queue Cycling

Wald¹,

Jaroš

Zellmann

2023

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…Our framework builds on small, self-contained path nodes that can be forwarded across the network. Each path contains a ray origin and direction, a throughput value [Boksansky and Marrs 2021], and the pixel ID to which it belongs, plus some bits to indicate whether a ray is a shadow ray, in medium, etc. To track already visited nodes we use a bit-field of either 8 or 64 bits depending on number of ranks; for more ranks we would use the technique described in Section 3.3.7.…”

Section: Rays Paths Hits and Ownership Masksmentioning

confidence: 99%