I’m completely speechless. This looks so terrible I thought it was a joke, but apparently Nvidia released these demos to impress people. DLSS 5 runs the entire game through an AI filter, making every character look like it’s running through an ultra realistic beauty filter.
The photo above is used as the promo image for the official blog post by the way. It completely ignores artistic intent and makes Grace’s face look “sexier” because apparently that’s what realism looks like now.
I wouldn’t be so baffled if this was some experimental setting they were testing, but they’re advertising this as the next gen DLSS. As in, this is their image of what the future of gaming should be. A massive F U to every artist in the industry. Well done, Nvidia.
Reading through the comments, this is definitely an unpopular opinion, but while I don’t love it, it does look more “realistic” to me, even though it definitely needs to look more like the source material.
Exactly. This is one of the few good uses for AI. It looks less like a beauty filter and more like a realism one. If it can make the rest of the game look as real, I’m all for this. I just hope it doesn’t require a GPU upgrade, because I was lucky enough to score a 4090 at launch for MSRP, and I plan on keeping it for at least another 6-8 years.
Brother, get your eyeballs checked. Nothing about that looks photoreal.
Yeah no at this point nuke the whole fucking company.
Someone reported your comment.
that’s glorification of violence and then there’s nuclear options…
You don’t have a pocket nuke, do you?
Can’t believe DF is so positive about this, just looks horrible. I’ve actually been quite positive about technologies like DLSS and FSR but this… no thanks.
They absolutely got a huge nvidia check for this one
Crazy to me to see people just now waking up to just how underinformed Digital Foundry really is to technical details, and how much they sold out.
Well, helps that you don’t need to know how to turn a computer on to see that they’re glazing some complete bullshit, just eyes.
This painfully true
Do you guys turn your graphics settings to lowest to make sure not to interfere with “artistic vision”? Or do you just not understand the technology at all?
The dev can literally control the vectors for DLSS 5 and test it for themselves until it looks how they like. If you don’t like modifying frames, you should disable all post-processing shaders and most modern anti-aliasing too.
Do you guys turn your graphics settings to lowest to make sure not to interfere with “artistic vision”?
This such a bad take and demonstrates a total disregard for the role of technical artists and design.
Anti-aliasing aims to overcome an inherent limitation a raster (of pixels) may present, in oder to bring the actual result closer to an artist’s vision. The same can be said for any other technique. The common denominator is, that they all attempt to bring a artistd vision for their work closer to reality.
DLSS5, on the other hand, brings the result closer to Nvidias vision of what “good” looks like. And I think it looks bland and badly photoshopped and I want to see more good looking games.
Not true at all. Have you not seen all the NES emulators that apply anti-aliasing or other modern graphics techniques to old NES games? Completely changes the dev’s vision. You need to disable them.
Or you can just realize that modern devs are working with FSR/DLSS and tweaking it how they want. Just like they been doing with anti-aliasing. They can literally mask portions of the scene so DLSS 5 doesn’t apply to it.
Now, many games they showcased likely didn’t put much effort into tweaking their performance for DLSS 5 since it’s new and they didn’t get the chance, but newer games will use it. It’s not like raytracing and pathtracing just enhance whatever GPU driver devs want it to. The devs can control it.
Isn’t DLSS by defintion always an AI slop filter?
Not really, DLSS mostly just reduces the resolution of a game and then upscales it back up. It does a pretty good job of making the game still look (almost) exactly the same. This, however, completely changes what you’re looking at.
DLSS is short for Deep Learning Super Sampling, it does the upscaling using deep learning, what people also call AI. The upscaler has to be trained on images. Depending on how you train it you either get something that looks almost exactly the same as the game at a higher resolution or you get AI slop.
I’m aware of how it works, but the results aren’t bad. Worst case scenario is you get some ghosting with DLSS, but it’s far from what I’d call AI slop.
But it literally follows the same process. Why is one slop, but not the other? You’re being hypocrital.
One is upscaling the image while preserving it as much as possible, the other is applying a filter to try and “enhance” it by drastically changing the image and ignoring artist’s intent. What’s hard to get?
This isn’t applying a filter, it’s
applyingrunning the image through a transformer network trained on advanced lighting methods like subsurface scattering to make materials more lifelike. It seems to change artistic intent quite a lot on these existing games, but frankly I’m excited to see what creators do with a game designed from the ground up to utilize AI-enhanced lighting. The DF video also states that this is an early preview (hence the dual 5090s) that is expected to change over time.it’s applying advanced lighting methods like subsurface scattering to make materials more lifelike.
It is not. It is approximating the results of training data consisting of output images that have been rendered with subsurface scattering. It isn’t actually running the subsurface scattering algorithm.
How is “upscaling while preserving it” not the exact same philosophy as “enhance by applying a filter?”
You just don’t like the specific filter, it’s very literally the same process.
… How if flying a spaceship different from driving a car? They’re both controlled applications of kinetic energy to move people or objects.
At the end of the day, it’s all a pile of transistors and the only thing that is of import is the intent behind usage.
In one case it’s saying you can use a neural net to take something rendered at resolution A/4 and make it visually indistinguishable from the same render at resolution A.
The other is rendering something and radically changing the artistic or visual style.Upsampling can be replicated within some margin by lowering framerate and letting the GPU work longer on each frame. It strives to restore detail left out from working quicker by guessing.
You cannot turn this feature off and get similar results by lowering the frame rate. It aims to add detail that was never present by guessing.Upsampling methods have been produced that don’t use neural networks. The differences in behavior are in the realm of efficiency, and in many cases you would be hard pressed to tell which is which. The neural network is an implementation detail.
In the other case, the changes are more broad than can be captured by non AI techniques easily. The generative capabilities are central to the feature.Process matters, but zooming out too far makes everything identical, and the intent matters too. “I want to see your art better” as opposed to “I want to make your art better”.
Current DLSS intent: We can only render this at like 720p with enough frames, so let’s do that and use AI anti-aliasing tricks so that when we present it at 4k, none of the jaggies are visible on-screen like they would be with raw 720p upscaling.
DLSS5 intent: Using our pile of stolen artwork neural net that we can now render at 60fps+ let’s “reimagine” the entire look of the game as we present it on-screen, even if it was already running at 4k just fine.
TLDR; How big the neuralnet is and what your train it for matters.
Because a pixelated circle being upscaled is a circle, but a pixelated circle being turned into a high definition pie is no longer a circle, and that’s especially problematic if the circle was just a cross hair or some other random circle like thing the AI thought was meant to be a pie.
Yes, both things are the same, but that’s like saying you had a tiny spider in your house and you were okay because it killed mosquitoes in your house, so you should be okay with having a colony of bats since they are also animals and eat mosquitoes. Yes, both are the same, but the scales and the amount of intrusion are completely different.
Not all answers are easy. This new dlss looks like it was trained on stolen work. Old dlss had a neutral network that was tuned before the plagiarism machine became popular.
Piracy is not stealing.
It is when it’s used by corporations for profit, IMO. Not for individual private enjoyment.
Hey at least it still runs on your machine.
DLSS 6 will run exclusively on the cloud because RTX 6 series won’t ship with any RAM or compute nodes since they all went into enterprise cards lol
actually helps out in the sports games. ea has the same lightning like they had in 2006, so this takes it up to speed to 2020. however real games gonna look arse. alao 2 x 5090 cards required, donno who the fuck is the demographic. also thanks for killing dual gpu support so you could bring a shit one back.
New Worms game looking hype though
The second one didn’t look too different from the original, but the first and third look so far from the source material, it’s jarring.
It feels like homogenization of art. The filter pays no attention to the original art style and just conforms it to nividias Ai style.
The devs can literally control the vectors used. They can even mask it to not apply to some things. Not sure how it calculates lighting though.
This looks so bad, like uncanny 3d porn bad.
Yup. You will soon pay premium for checks notes actual footage of people enjoying fucking?!
oof the dislike ratio on this video is brutal. 23k likes but 49k dislikes, at least according to PiePipe.
as it should be. DF has instantly lost all the credibility they worked so hard to earn for so many years.
Insanely impressive that it’s able to do that in real time.
