"o tom že DLSS není forma upscalingu atd… ten už se stal legendární."
Muzes to opakovat do zblbnuti a stejne se z toho nestane pravda :)
"In previous years, NVIDIA’s founder and CEO, Jensen Huang, liked to show a Deep Learning neural network demo where artificial intelligence would decide whether a photo of a pet was showing a cat or a dog. As the machine got more images and made more correct picks it would get better and better, before eventually picking the correct answer each time.
Using that basis for machine learning, we’ve created a new rendering technique called Deep Learning Super Sampling (DLSS)."
https://www.nvidia.com/en-us/geforce/news/graphics-reinvented-new-technologies-in-rtx-graphics-cards/#dlss
To je princip autoencoderu. Nauci se povahu objektu (jestli vidi, psa, kocku, auto, atd.) a jsou schopne je poznat ve scene a na zaklade toho doplnit chybejici detaily (samply). Takze zadnej spatial filtr, sorry, ale ne sorry :D