Memory-Hazard-Aware K-Buffer Algorithm for Order-Independent Transparency Rendering

The (k)-buffer algorithm is an efficient GPU-based fragment level sorting algorithm for rendering transparent surfaces. Because of the inherent massive parallelism of GPU stream processors, this algorithm suffers serious read-after-write memory hazards now. In this paper, we introduce an improved (k...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on visualization and computer graphics Jg. 20; H. 2; S. 238 - 248
1. Verfasser: Zhang, Nan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.02.2014
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Algorithms
Computer graphics
Error analysis
error checking code
graphics processor
Hazards
memory hazard
Processors
Rendering
Rendering (computer graphics)
Sorting algorithms
Streams
Transparency rendering
visibility sort
Visualization
Volume measurements
volume visualization
ISSN:1077-2626, 1941-0506, 1941-0506
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The (k)-buffer algorithm is an efficient GPU-based fragment level sorting algorithm for rendering transparent surfaces. Because of the inherent massive parallelism of GPU stream processors, this algorithm suffers serious read-after-write memory hazards now. In this paper, we introduce an improved (k)-buffer algorithm with error correction coding to combat memory hazards. Our algorithm results in significantly reduced artifacts. While preserving all the merits of the original algorithm, it requires merely OpenGL 3.x support from the GPU, instead of the atomic operations appearing only in the latest OpenGL 4.2 standard. Our algorithm is simple to implement and efficient in performance. Future GPU support for improving this algorithm is also proposed.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1077-2626
1941-0506
1941-0506
DOI:10.1109/TVCG.2013.72