Recently, PC gamers gained access to the NVIDIA DLSS 5 library. Modders were able to extract the necessary files from NBA 2K27. Soon, enthusiasts began using the technology in a variety of games, including Kingdom Come: Deliverance II.
The reaction to such experiments varies: some are pleased with the image, while others find what they see to be a horror and "neural slop."
Daniel Vavra, the creator of Kingdom Come: Deliverance II, noticed what was happening and decided to experiment himself. He liked it and began to defend the technology.
He suggested looking at an example of a character's head model — NVIDIA technology helps improve the image, but, Vavra explains, it does not change the geometry:
Some people have literally gone insane. On the left side of the picture is a head spoiled by a very limited game engine lighting system that cannot correctly display shadows, skin texture, and normal map details. On the right is the SAME MODEL WITHOUT ANY GEOMETRY CHANGES, which looks exactly like a render from ZBrush — where the 3D artist modeled it — only with better lighting, shadows, skin texture, hair, and, consequently, details. This is possible thanks to neural network rendering in DLSS 5. And then some random crazy person on the internet writes: "THIS IS NEURAL SLOP! I WANT THE ORIGINAL CREATIVE VISION."
Daniel Vavra said that the head models were 3D scanned (using photogrammetry) from real people and then refined by artists:
I am the creative director of the game. This is our original creative vision, and [the model] looks like static renders of models from ZBrush when we created them. Or like the real people [from whom the scans were made]. The absence of helmet shadows on faces and distorted normal map rendering WERE NOT OUR ORIGINAL CREATIVE INTENTION. People are just insane.