The Open, Scalable, and Portable Ray Tracing Engine

Render massive scenes interactively

Disney’s Moana Island Scene: over 15 billion instanced primitives+volume at interactive rates

Photorealism

Physically-based shading and lighting

Feature rich, large-scale volume visualization

Richtmyer–meshkov volume shown with shadows and ambient occlusion

USD Hydra Integration

Gramophone rendered in Pixar’s usdview

Integrations in VTK, ParaView, VisIt, and more

DKRZ weather data rendered with shadows inside ParaView

Directly render unstructured and AMR volumes

AMR volume rendered without resampling inside ParaView

Interactive CPU Rendering

OSPRay features scalable CPU rendering capabilities geared toward Scientific Visualization applications. Advanced shading effects such as ambient occlusion, shadows, and transparency can be rendered interactively to enable new insights into huge data.

Global Illumination

OSPRay includes a path tracer capable of interactively rendering photorealistic global illumination with phyiscally-based materials.

Volume Rendering

OSPRay supports high-fidelity, interactive direct volume rendering with a number of state of the art visualization features.

MPI Distributed

Run on large scale distributed-memory systems with high-performance MPI backends to both decrease render time and increase total scene size.

Industry Adoption

OSPRay is directly integrated into ParaView 5.x. Implementations for VisIt, VMD, and other popular tools have also freely available.

Open Source

OSPRay is Open Sourced under the Apache 2.0 license.

OSPRay Overview

Intel OSPRay is an open source, scalable, and portable ray tracing engine for high-performance, high-fidelity visualization on Intel Architecture CPUs. OSPRay is part of the Intel oneAPI Rendering Toolkit and is released under the permissive Apache 2.0 license.

The purpose of OSPRay is to provide an open, powerful, and easy-to-use rendering library that allows one to easily build applications that use ray tracing based rendering for interactive applications (including both surface- and volume-based visualizations). OSPRay is completely CPU-based, and runs on anything from laptops, to workstations, to compute nodes in HPC systems.

OSPRay internally builds on top of Intel Embree and ISPC (Intel SPMD Program Compiler), and fully exploits modern instruction sets like Intel SSE4, AVX, AVX2, and AVX-512 to achieve high rendering performance, thus a CPU with support for at least SSE4.1 is required to run OSPRay.

OSPRay Support and Contact

OSPRay is under active development, and though we do our best to guarantee stable release versions a certain number of bugs, as-yet-missing features, inconsistencies, or any other issues are still possible. Should you find any such issues please report them immediately via OSPRay’s GitHub Issue Tracker (or, if you should happen to have a fix for it,you can also send us a pull request); for missing features please contact us via email at ospray@googlegroups.com.

To receive release announcements simply “Watch” the OSPRay repository on GitHub.

Changes in v2.5.0:

Add native support for cones or cylinders with curves geometry of type OSP_DISJOINT, requiring minimum version 3.12.0 of Embree
Replaced OSPRay’s internal implementation of round linear curves by Embree’s native implementation. Internal surfaces at joints are now correctly removed, leading to higher quality renderings with transparency, at the cost of intersection performance
SciVis renderer improvements:
- Colored transparency, colored shadows
- Light sources are visible including HDRI Light environment map
The MPI module is now distributed as part of OSPRay in the modules directory
- The socket-based communication layer has been removed
Add intensityQuantity parameter to light sources to control the interpretation and convertion of the intensity into a radiative quantity
OSPRay now requires minimum Open VKL v0.12.0 to bring the following improvements:
- Better default sampling rate for scaled volumes, improving performance
- Higher robustness for axis-aligned rays
Removed limit on the number of volumes (both overlapped and separate) that a ray can intersect while rendering. Now it is limited by available memory only.
Move to OIDN v1.3.0 to bring the following improvements:
- Improved denoising quality (sharpness of fine details, fewer noisy artifacts)
- Slightly improved performance and lower memory consumption
Both geometric and volumetric models can now have their child geometry/volume objects rebound using an object parameter
Fix light leaking artifacts at poles of HDRI (and Sun-Sky) light
Add sRGB conversion to ospExamples such that the color of the widget for backgroundColor actually matches
Dropping support for MSVC14, new minimum compiler on Windows is MSVC15 (Visual Studio 2017)

Changes in v2.4.0:

The pathtracer optionally allows for alpha blending even if the background is seen through refractive objects like glass, by enabling backgroundRefraction
OSPRay now requires minimum Open VKL v0.11.0 to bring the following improvements:
- Improved rendering performance of VDB volumes
- Added support for configurable iterator depth via the maxIteratorDepth parameters for unstructured and particle volumes, improved performance
- Added support for filter modes for structured volumes (regular and spherical)
Expose parameter horizonExtension of Sun-sky light, which extends the sky dome by stretching the horizon over the lower hemisphere
Optimize handling of geometry lights by the pathtracer
The optional denoiser image operation now respects frame cancellation, requiring Intel® Open Image Denoise with minimum version 1.2.3
Fixed normals of (transformed) isosurfaces
Robust calculation of normals of boxes geometry
Clipping geometry is now working correctly with map_maxDepth renderer parameter
Using materials in a renderer with a mismatched renderer_type no longer causes crashes while rendering
Significant improvements have been made to loading performance in the MPI Offload device. Applications which make large numbers of API calls or create many smaller geometries or volumes should see substantial load time improvements.
MPI module compacts strided data arrays on the app rank before sending

For the complete history of changes have a look at the CHANGELOG.