Shortly after the Khronos Group announced official ray tracing extensions for the Vulkan graphics API, Nvidia had a raytracing announcement of its own. Jensen and Friends, who are best known for locking down graphics technologies with proprietary code and arbitrary platform restrictions, surprised everyone by releasing their RTX Global Illumination SDK 1.0 to the community at large as an open-source solution. The Global Illumination SDK allows developers to deploy ray-traced global illumination effects on any DirectX Raytracing (DXR) compliant graphics card. This means that RTXGI, the highlight feature in a number of RTX-exclusive titles like Metro: Exodus and Control, could be coming to Intel Xe and AMD’s RDNA2 cards.
How exactly does this work, though, and what are the implications for graphics cards that don’t have dedicated ray-tracing hardware? Speaking to Guru3D, NVIDIA’s Tony Tamasi had this to say:
Essentially, the RTXGI SDK allows developers to deploy scalable solutions that enable limited ray-traced global illumination even on graphics cards with no support for hardware-accelerated ray tracing. The effect quality can then be ramped up with higher-end hardware with more dedicated units.
The implications here are major. With a number of other players jumping aboard the ray-tracing bandwagon, it appears that NVIDIA is finally conceding exclusivity. Considering that the ninth-gen consoles all have RDNA2 hardware ray-tracing tech built-in, it’d be interesting to see if upcoming RTX titles like Cyberpunk 2077 end up shipping with hardware-agnostic ray-tracing effects.