How does Gelato differ from the native renderer on the GPU?
The native renderer on the GPU is focused on rendering to the screen and is optimized to maintain a high-frame rate at the expense of quality. Gelato is exactly the opposite; it never compromises on quality. Gelato is designed from the ground up to exploit the massive floating point capability of NVIDIA GPUs, avoiding artifacts that are acceptable in real-time rendering, but are not in high-quality, offline rendering. Gelato also has a higher-level interface than the native renderer on the GPU, affording more flexibility and the ability to use advanced techniques, like global illumination, that the hardware is not yet capable of.
