- Homepage: mitsuba
- Current release: mitsuba renderer 0.5.0 from February 2014
- Standalone renderer, has its own XML format but there exist a tool to convert COLLADA files to the proprietary format. There are mitsuba exporters for Maya, Blender, Softimage and Rhino.
The renderer is written by Wenzel Jakob. Mitsuba contains biased as well as unbiased techniques. It offers as well an interesting approach to volume rendering where grid cells in a file are directly mapped into memory.
Mitsuba now offers a great python integration. Multichannel rendering (AOV's) are now supported. The standalone now has an interactive preview.
This experimental renderer supports a bundle of different rendering techniques. That means you can create scenes and render with different approaches to see what fits best. The UI has now realtime feedback, have a look at the impressive video. There are different sensors (camera sensors) available. Additionally a skylight emitter illuminates the scene and the renderer can use large textures. It's a great release. Unfortunatly there is no api for 3rd party DCC plugin developers, what means there is no easy way to implement it into an existing 3d application like maya.
Last update: 01.2021