DLSS5...aka Ai slop?

[5n0man]

Distopian future in Exchange for GTA World for the win

Every other day I see a story about executives and CEOs of these companies talking about plans for a future where they could have a monopoly on an a.i. driven economy.

They see the artist and the programmers as middle men now, I dont know how anyone could be excited for this technology.

 

[Rain]

the way you put it down, make sense. I can see why many people especially PC gamers are upsetting this, against this.

Idk. I wouldn't give most gamers credit for having this type of nuanced reasoning for being opposed to this. I think they hear/see "AI" and their first thought is to reject it off principle, which I get, but...

If you put these same DLSS5 images side to side, and say it's a comparison between PS5 Pro vs base XSX, most of them would be remarking about how much of a generational leap it is and how far we've came. Say it's AI and everyone acts like they don't like what they're seeing

We can keep trying to reject AI or people can get on board and control how its used. Either way it's gonna be apart of gaming one way or another. And if using AI can speed up development timelines, fix limitations like textures/assets reuse, etc, then it has a place in gaming whether people like it or not.

 

[Tim Lord]

Theres simply no argument against this.

These gifs alone show you that the faces remain the same, just rendered at a vastly higher level of quality. And it does most of it with just lighting. Its not changing anybodys face. They are clearly the same people. The Ai is being trained by the game itself, not some random person online.... So the characters will still look like how the artists designed them.

Look at the last gif, you can't even see his eyes without DLSS 5. He has eyes now.

I think people who dislike it are focusing way too much on the faces.

They're missing the bigger picture ... the potential it has for actual gameplay and environment visuals...

1. Faces really only matter during cutscenes, not moment-to-moment gameplay.
2. Players spend about 95% of their time traversing the world, interacting with the environment.

That's where this tech really shines. The environmental upgrades I've seen in clips are fantastic.

This tech is brand new, still a baby in infancy. And we're already getting results like this, from studios that aren't even top tier.

Imagine when CDPR, Game Science, Pearl Abyss, R*, get their hands on this? It's going to be absolute insanity.

I am more excited for the potential.

This has the potential to speed up development in the long run. Because inevitably, AI will make spending years optimizing a game a thing of the past. Development becomes faster, iteration becomes easier, and teams can spend less time fighting technical limitations and more time building bigger worlds, deeper systems, and more ambitious ideas.

This is an in-game photo with no HUD with DLSS5.

Look at those visuals , just look at them. Her clothing looks life like, draw distance insane, level of detail insane

I can’t speak to the first two images you shared as they were NVIDIA tech demos made specifically to show off this tech. I don’t have a reference point for it so no comment on those.

For the rest, I can share some thoughts. One of the things that DLSS5 is supposed to address is lighting.

For the non DLSS5 Oblivion character, he has eyes, it’s just that the 3D model’s eye socket area (the geometric mesh) is deep :pause: and embedded in his face. So his face is capturing the light and the mesh is creating shadows in the eye area making it hard to see. In real life we would see someone’s eyes since light would still bounce on their face and illuminate the eye area. If the game was path traced or if the ray tracing had an extra bounce it would resolve the dark eye area. One thing that is apparent is that it is very easy to tell where the sun is within the non DLSS5 version.

The AC:Shadows screen shot is a lot more cut and dry. I can fully understand people preferring the look of the DLSS5 version, but once again, it is allegedly supposed to be creating better lighting. This character is outside and in the sun. If you were to solely look at the DLSS5 image, where would you place the position of the sun (which is supposed to be the source of light within this scene)?

On the ray traced version we can see a warm orange sun tint on the characters back (indicating the suns direction) which then shadows her right knee as it is being blocked by the sun.

In the DLSS5 version the scene is uniformly lit and a warmer toned orange sun turns into whitish-blue flood light. Her body is uniformly lit and in the brief video clip online, the clouds do not cast shadows. So its lighting/shadows can be viewed as less accurate and even directionless and the tone of the scene changed based on the ML models output.

Yes this is all early but we can only judge what we have been presented. Plus this has been presented publicly so it’s fair game to criticize it, and, it’s a tech demo which is usually the best case scenarios when showing off new tech.

 

[VegetasHairline]

I can’t speak to the first two images you shared as they were NVIDIA tech demos made specifically to show off this tech. I don’t have a reference point for it so no comment on those.

For the rest, I can share some thoughts. One of the things that DLSS5 is supposed to address is lighting.

For the non DLSS5 Oblivion character, he has eyes, it’s just that the 3D model’s eye socket area (the geometric mesh) is deep :pause: and embedded in his face. So his face is capturing the light and the mesh is creating shadows in the eye area making it hard to see. In real life we would see someone’s eyes since light would still bounce on their face and illuminate the eye area. If the game was path traced or if the ray tracing had an extra bounce it would resolve the dark eye area. One thing that is apparent is that it is very easy to tell where the sun is within the non DLSS5 version.

The AC:Shadows screen shot is a lot more cut and dry. I can fully understand people preferring the look of the DLSS5 version, but once again, it is allegedly supposed to be creating better lighting. This character is outside and in the sun. If you were to solely look at the DLSS5 image, where would you place the position of the sun (which is supposed to be the source of light within this scene)?

On the ray traced version we can see a warm orange sun tint on the characters back (indicating the suns direction) which then shadows her right knee as it is being blocked by the sun.

In the DLSS5 version the scene is uniformly lit and a warmer toned orange sun turns into whitish-blue flood light. Her body is uniformly lit and in the brief video clip online, the clouds do not cast shadows. So its lighting/shadows can be viewed as less accurate and even directionless and the tone of the scene changed based on the ML models output.

Yes this is all early but we can only judge what we have been presented. Plus this has been presented publicly so it’s fair game to criticize it, and, it’s a tech demo which is usually the best case scenarios when showing off new tech.

Whose alt is this?

 

[Rain]

Whose alt is this?

Idk, but the thought crossed my mind that these posts are AI which would make this shyt funny AF

 

[Tim Lord]

Well, I guess that is enough for today, back to the shadows.

People can just admit that they like overprocessed, oversharpened and oversaturated A.I. generated shyt. If that’s the case, there is not much to discuss.

 

[Lord Beasley]

In some games it looks better and in some games it looks worse.




Now on to the important shyt. Can someone run this tech with Dead Or Alive Beach Volleyball so we can get a true representation of its capabilities.

Show me games where it looks worse

 

[Tair]

Theres simply no argument against this.

These gifs alone show you that the faces remain the same, just rendered at a vastly higher level of quality. And it does most of it with just lighting. Its not changing anybodys face. They are clearly the same people. The Ai is being trained by the game itself, not some random person online.... So the characters will still look like how the artists designed them.

Look at the last gif, you can't even see his eyes without DLSS 5. He has eyes now.

I think people who dislike it are focusing way too much on the faces.

They're missing the bigger picture ... the potential it has for actual gameplay and environment visuals...

1. Faces really only matter during cutscenes, not moment-to-moment gameplay.
2. Players spend about 95% of their time traversing the world, interacting with the environment.

That's where this tech really shines. The environmental upgrades I've seen in clips are fantastic.

This tech is brand new, still a baby in infancy. And we're already getting results like this, from studios that aren't even top tier.

Imagine when CDPR, Game Science, Pearl Abyss, R*, get their hands on this? It's going to be absolute insanity.

I am more excited for the potential.

This has the potential to speed up development in the long run. Because inevitably, AI will make spending years optimizing a game a thing of the past. Development becomes faster, iteration becomes easier, and teams can spend less time fighting technical limitations and more time building bigger worlds, deeper systems, and more ambitious ideas.

This is an in-game photo with no HUD with DLSS5.

Look at those visuals , just look at them. Her clothing looks life like, draw distance insane, level of detail insane

you see the vision.

Dap+Rep

 

[42 Monks]

Well, I guess that is enough for today, back to the shadows.

People can just admit that they like overprocessed, oversharpened and oversaturated A.I. generated shyt. If that’s the case, there is not much to discuss.

WHO ARE YOU

 

[TrifeGod]

I can’t wait to inject this shyt in sonic and knuckles fall of this year will be july again lol we about to see some wild shyt