Lors de sa récente conférence GTC 2026, Nvidia a levé le voile sur le DLSS 5 (Deep Learning Super Sampling), marquant une rupture majeure dans l’évolution du rendu graphique. Si les versions précédentes se concentraient...
Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.
Howdy howdy ho and welcome to fantasy fanfellas. I'm Hayden producer of the fantasy fangirls podcast and your resident lover of all things
Sanderson and I'm Steven your bookish internet goofball, but you can call me the smash daddy and we are currently deep diving
Brandon Sanderson's fantasy epic Missborn, but here's the catch Steven here has not read Missborn before that's right
Hey, so each week you'll get my unfiltered raw reactions to every single
And along the way we'll do character deep dives magic explainers and Steven will even try to guess what's next spoiler alert
He'll be wrong news flash. I'm never wrong episodes come out every Wednesday and you can find fantasy fanfellas wherever you get your podcasts
There are 48 hours Nvidia at the Veloil on the DLS C5
What is it? It's a technology which is a major rupture, we can say, in the evolution of the graphic
Why a rupture? Well, because the previous versions focus on the optimization of performance and fluidity
especially via the scaling and the generation of images, while this new iteration re-invents this aesthetic
Thanks to LIA, thanks to the artificial intelligence generated
And so this DLS C5 does not contain more simply moving the holes between the pixels
In it uses a neural rendering model which was trained to understand the semantic scenes
Very concretely, LIA analyzes the textures, the skin, the hair and the lighting to reinterpret them in real life
The objective is to reach a photorealism until reserved in the films
LIA simulates with an unprecedented precision the diffusion of the light on the materials
All of that is pretty, but the spectacular aspect of the demo that we were presented, especially on Resident Evil or Starfield, has not received an anime
That's what we can say
The first images even have a polemic life
So why is the main grieff here concerns the nature of the artistic direction of the game?
Well, there is a lot of improvement at this point, the faces, LIA of Nvidia, while it is smoothed to modify the traits
And to add the macchiage between gimmicks, there, there, there, there was no
There is also a certain effect of FITRA Instagram, which is judged by many artificial
Even if paradoxically, it is close to reality
And for many observers and artists, this technology, it creates the original intention of creators to offer an aesthetic uniformity
The characters, they lose their soul
So this Levet de Bouquillier, Nvidia, it is rather reassuring, the firm, has linked that the DSC-5 is an entirely parametrable tool
For example, the studio of development will keep the control on the intensity of the effects
And they will be able to choose not to apply to some elements
So if the DSC-5 proves the technological domination of Nvidia, there is a ethical and artistic question
The following, the video game, must it be absolutely a realism dictated by LIA?
Or, against it, stay a kind of sanctuary, in which the human vision, which is, in itself, a perfect, staying singular
The answer will come to some of the first things that are concrete in the games, planned for the end of the year