TauBench v1.1 is a permissively licensed (CC-BY-NC-SA 4.0) dynamic benchmark for graphics rendering, especially for benchmarking temporal reuse methods. It contains two scenes in binary glTF 2.0 format: EternalValleyFPS and EternalValleyVR.

• EternalValleyFPS represents a typical interactive multiplayer game scenario with dynamically changing conditions and geometry animations.
• EternalValleyVR adds rapid camera motion caused by head-mounted displays popular with virtual reality applications.

Download TauBench at https://doi.org/10.5281/zenodo.5729573. It contains the EternalValleyFPS and EternalValleyVR glTF files (.glb), and the example rendered videos as .mp4 (in total 2.0 GB compressed).

• Reference PNG frames (both 1080p and 2160p resolutions, rendered with Tauray at 16384 spp) are available at https://zenodo.org/record/7907164 (in total 18.8 GB compressed).

• v1.1 (May 2023) - Technical notes (PDF)
• Instanced grass and pebbles to reduce the .glb file sizes to about one third of the original.
• Removed point lights that were fully occluded (= point lights inside the moving grenades).
• Snapped animation key frames to integer frames to fix blender export issues. This naturally changes when some of the animations occur.
• Fixed minor colour artifacts on the torches.
• Added higher-quality 2160p example videos.
• Added 1080p and 2160p reference frames as PNG files, rendered with Tauray at 16384 spp. Not included in the main release, but uploaded as a separate archive.

• v1.0 (February 2022)
• Initial release

Import one of the .glb files in your 3D modelling software, renderer or glTF viewer (for example the newest version of blender).

• Changing the default frame rate from 24 fps to 60 fps in blender: Properties - Output Properties - Frame Rate

• Properties:
• Length: ~6 seconds (at 60 fps)
• File format: glTF 2.0 (.gltf)
• Triangles: 19.0M
• Meshes: 264 in EternalValleyFPS, 262 in EternalValleyVR
• Textures: 133 in EternalValleyFPS, 129 in EternalValleyVR
• Static objects: 48251 in EternalValleyFPS, 48255 in EternalValleyVR
• Dynamic rigid body objects: 491 in EternalValleyFPS, 477 in EternalValleyVR
• Armatures: 32, with up to 52 animated bones
• Point lights: 62
• Each animation frame has Translation, Rotation and Scale channels
• Material description:
• Most objects have 3 textures: BaseColor, MetallicRoughness and Normal
• Textures are png encoded
• BaseColor:
• RGB channels: BaseColor value
• Alpha channel: Transparency
• MetallicRoughness:
• Green: Metallic
• Blue: Roughness

Many graphics rendering algorithms used in both real time games and virtual reality applications can get performance boosts by reusing previous computations. However, the temporal reuse based algorithms are typically measured using trivial benchmarks with very limited dynamic features. To this end, we present two new benchmarks that stress temporal reuse algorithms: EternalValleyVR and EternalValleyFPS. These datasets represent scenarios that are common contexts for temporal methods: EternalValleyFPS represents a typical interactive multiplayer game scenario with dynamically changing lighting conditions and geometry animations. EternalValleyVR adds rapid camera motion caused by head-mounted displays popular with virtual reality applications. In order to systematically assess the quality of the proposed benchmarks in reuse algorithm stress testing, we identify common input features used in state-of-the-art reuse algorithms and propose metrics that quantify changes in the temporally interesting features. Cameras in the proposed benchmarks rotate on average 18.5x more per frame compared to the popular NVidia ORCA datasets, which results in 51x more pixels introduced each frame. In addition to the camera activity, we compare the number of low confidence pixels. We show that the proposed datasets have 1.6x less pixel reuse opportunities by changes in pixels’ world positions, and 3.5x higher direct radiance discard rate.

@inproceedings{alanko2022taubench,
	title={TauBench: Dynamic Benchmark for Graphics Rendering},
	author={Alanko, Joel and M\"{a}kitalo, Markku and J\"{a}\"{a}skel\"{a}inen, Pekka},
	booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - GRAPP},
	year={2022},
	pages={172--179},
	publisher={SciTePress},
	organization={INSTICC},
	doi={10.5220/0010819200003124}
}

Preprint of the v1.0 paper

Technical notes of the v1.1 release

Supplementary material (v1.0)

Top: EternalValleyFPS, bottom: EternalValleyVR. Larger images here.