• Corpus ID: 40711158

Optimizing Draw Call Batching Using Transient Data-Guided Texture Atlases

@inproceedings{Dortmont2017OptimizingDC,
  title={Optimizing Draw Call Batching Using Transient Data-Guided Texture Atlases},
  author={J.A.L.T. van Dortmont},
  year={2017}
}
Rendering a large number of 2D textures in real-time requires reducing the overhead of a large number of draw calls on the CPU caused by binding different textures when drawing. Texture atlases are used to avoid switching textures by packing textures into one larger texture before rendering. Graphics hardware APIs limit the size of a texture, so textures need to be partitioned into multiple atlases. Unfortunately, composing textures into atlases is performed manually by developers or artists… 
osiris.library.uu.nl

References

SHOWING 1-10 OF 26 REFERENCES
Geometry Batching using Texture-arrays
TLDR
This paper presents a pre-processing approach and rendering technique that overcomes limitations by further grouping textures or texture atlases and thus enables the creation of larger geometry batches, based on texture arrays in combination with an additional indexing schema that is evaluated at run-time using shader programs.
Advanced virtual texture topics
TLDR
This chapter is discussing technical implications on engine design due to virtual textures use, content creation issues, results, performance and image quality, and several practical examples to highlight the challenges and to offer solutions.
Virtual Texture Mapping 101
TLDR
This chapter investigates how to implement a fully functional VTM system, a technique to reduce the amount of graphics memory required for textures to a point where it is only dependent on the screen resolution.
Prefetching in a texture cache architecture
TLDR
This paper introduces a prefetching texture cache architecture designed to take advantage of the access characteristics of texture mapping, and demonstrates that even in the presence of a high-latency memory system, this architecture can attain at least 97% of the performance of a zerolatency memory systems.
Accelerating Virtual Texturing Using CUDA
TLDR
Virtual texturing offers generalized texture streaming that works on arbitrary geometry & unwraps and dynamic streaming & compression of tiles needed for parallel processing.
The clipmap: a virtual mipmap
TLDR
The clipmap, a dynamic texture representation that efficiently caches textures of arbitrarily large size in a finite amount of physical memory for rendering at real-time rates, is described and ways that other rendering systems may exploit the concepts underlying clipmaps to solve related problems are suggested.
Hardware-Independent Clipmapping
TLDR
A technique for efficient management of large textures and its real-time application to geometric models is presented, inspired by the clipmap idea, that caches in video memory a subset of the texture mipmap pyramid.
Real-Time DXT Compression
TLDR
The DXT compressor described in this paper is optimized using the Intel Multi Media Extensions and the Intel Streaming SIMD Extensions 2.0 which allow textures to be compressed real-time which is useful for textures created procedurally at run time or textures streamed from a different format.
Pyramidal parametrics
TLDR
This paper advances a “pyramidal parametric” prefiltering and sampling geometry which minimizes aliasing effects and assures continuity within and between target images.
Balanced Graph Partitioning
TLDR
A bicriteria approximation algorithm that for any constant ν > 1 runs in polynomial time and guarantees an approximation ratio of O(log1.5n) (for a precise statement of the main result see Theorem 6).
...
1
2
3
...