NVIDIA had earlier described DLSS 5 in a way that suggested a deeper understanding of the scene. The follow-up answers paint a narrower picture. When asked whether the model reads PBR (Physically Based Rendering) properties from the engine, NVIDIA said: “DLSS 5 only takes the rendered frame and: “DLSS 5 only takes the rendered frame and motion vectors as inputs. Materials are inferred from the rendered frame.” In other words, the model is not reading metallic, roughness, normal maps, or other underlying material properties directly.
That may explain why some preview images raised concerns. In one example, a character appears to gain hair detail in an area where it was not visible before. In another, facial details appear altered enough (like the nose) to raise questions about whether the model is changing the look of the character rather than only improving lighting. NVIDIA’s response was that “the underlying geometry is unchanged,”
This was intuitive and obvious to anyone paying attention to AI and who knows video game engines. Nvidia trying to imply anything else is really shitty.
When it comes to generating lighting and atmospheric and dynamic surface effects in real time, calculating each mesh or surface in the scene instead of one big collective pass would be even more demanding. And will likely be beyond their capabilities for a very long time.