"Well, first of all, they're completely wrong," Huang said in response.
"The reason for that is because, as I have explained very carefully,
DLSS 5 fuses controllability of the of geometry and textures and
everything about the game with generative AI," Huang continued.
He added that developers
can still "fine-tune the generative AI" to make it match their style,
adding that DLSS 5 adds generative capability to the existing geometry
of the game, but that it "doesn't change the artistic control."
"It’s not post-processing, it’s not post-processing at the frame level, it’s generative control at the geometry level," he said.
Huang
also said that developers can try the tool and see how they want to use
it, suggesting that it's up to a developer to try to make a "toon
shader" or see if the game should be "made of glass."
"All of
that is in the control — direct control — of the game developer," he
said. This is very different than generative AI; it’s content-control
generative AI. That’s why we call it neural rendering."
So setting aside the fact that he doesn't know how his own technology works or games are made... If it did control and generate things every frame at the geometry level that would be impressive from a technical standpoint. However similar questions come to mind as any other art: how do you keep it consistent and prevent it from turning to mush and/or hallucinating? Games aren't movies, players will get themselves into places they aren't supposed to or how do you deal with a non-linear open world? How can you make sure that the game will be the same experience from person to person, let alone the same person if they backtrack?
Other than curiosity and the continuing sale of more gpu power... which no one can afford and is probably a push to get people onto their game streaming service... I don't see the point of this tech. It's like we forgot you can lower your settings to get more frames and for some people they may never even notice the settings between medium and high fidelity on some games.
Community: Nvidia, how are scrambled eggs made? Nvidia: Scrambled eggs are made from wheat, and the hens have full control over the result. Community: Really, isn’t it made from eggs? Nvidia: It is made from eggs, but wheat is the main input for the eggs, wheat thereby acts as a control plane for making the scrambled eggs. The hens are also left to fine tune and control own style of the final flavor of the scrambled eggs.
Replies
Huang also said that developers can try the tool and see how they want to use it, suggesting that it's up to a developer to try to make a "toon shader" or see if the game should be "made of glass."
"All of that is in the control — direct control — of the game developer," he said. This is very different than generative AI; it’s content-control generative AI. That’s why we call it neural rendering."
Community: Nvidia, how are scrambled eggs made?
Nvidia: Scrambled eggs are made from wheat, and the hens have full control over the result.
Community: Really, isn’t it made from eggs?
Nvidia: It is made from eggs, but wheat is the main input for the eggs, wheat thereby acts as a control plane for making the scrambled eggs. The hens are also left to fine tune and control own style of the final flavor of the scrambled eggs.