• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AMD Doesn't Believe in NVIDIA's DLSS, Stands for Open SMAA and TAA Solutions

The entire money and time (even if it isn't much) that nvidia is spending on DLSS is a waste.
They should entirely focus on the Ray-Tracing implementation instead, Metro is a nice showcase for them, after I saw the Ray Traced global illumination in metro exodus it really made me think that Ray Tracing has the potential to improve things in games, Cyberpunk 2077 is a good opportunity for nvidia, a game that most enthusiast PC gamers actually care about, they have a lot of time left to properly implement Ray Tracing in it and maybe make the game a true showcase for their next Gen GPUs.

The current problem with raytracing is, that you can just use it as a cherry on top. The game still has to work without it. There is still no game which has complete lighting only through raytracing. I think you need a completely different game engine to do that. And then it would not work on normal GPUs any longer at sufficient performance. Even Nvidia has to think about that since they just released there GTX 1660ti.

And i think the pictures are labeled wrong as well. But we need more information about what resolution they were taken, what is the resolution DLSS scales up from (and fills out some details). And what the exact TAA and SMAA settings are.
 
i think the overall image output of DLSS has too much noise
 
taa is very blurred, i do prefer dlss on those images above, night and day compared. What the hell amd is smoking saying taa is better than dlss.

Look at dlss elsewhere, then it's opposite :p
 
running the 3Dmark comparison both runs look fine to me the only difference i see is one runs at 35 fps and the other at 50 fps that is quite a difference..

looking at the comparisons images in this thread the ones labeled DLSS look far better.. okay the DLSS knockers think they are labeled wrong.. maybe they are who knows.. he he

trog
 
This gimmick is just useless, for one simple reason you can't use it freely on every game, as an example now I only play one game BF1 and it doesn't support DLSS so even with a 2080ti I'd be stuck using normal AA solutions.
I think nVidia should have used their silicon budget on much more Cuda cores, and find a solution like a xbox 360 promised to offer basically free MSAA X2 or more.
 
I expected AMD to take the open source path and that sounds like it is the case (if they pursue it at all). Judging by this thread, DLSS is on the path to being yet another HairWorks: a few games will use it because NVIDIA paid them to. After that, no one supports it.
 
proprietary solution? nvidia made those all the time. for example TXAA and MFAA that only works on their GPU. AMD never make a single noise about those before. why suddenly they care when nvidia try to push DLSS? the issue probably is less about being proprietary or open. the way i see it some people like it and some are not. more or less the same how it was with FXAA before where some people don't like the blurring side effect and some other rather have FXAA than not having AA at all. AMD most likely care because DLSS can uplift the performance quite significantly regardless of the quality image. if DLSS can give nvidia performance advantage (when AMD can only use TAA) they probably want the public to know that the comparison is not really a fair one because of image quality.

personally i think AMD probably also want to offer something similar to DLSS (hence the talk about to use Direct ML before) but stuff like DLSS is not just simply "inject" the AA into the game. DLSS need the image to be trained using ML first. for nvidia this training cost is something they wiling to shoulder themselves instead of passing them to game developer. will AMD wiling to do the same? some people said this latest effort from nvidia is just wasting money but that's simply how they roll. they try to push something and when it does not work for them they just move on.

This gimmick is just useless, for one simple reason you can't use it freely on every game, as an example now I only play one game BF1 and it doesn't support DLSS so even with a 2080ti I'd be stuck using normal AA solutions.
I think nVidia should have used their silicon budget on much more Cuda cores, and find a solution like a xbox 360 promised to offer basically free MSAA X2 or more.

they can add more CUDA cores but doing so they will face another problem. the glimpse of that problem already here with RTX2080Ti.
 
Because every frame of a benchmark can be optimized for. Out in the wild, you are more likely to see those pictures actually reversed as far as image quality, based on what I have read so far.

Then tat would be Nvidia Falsely advertising and manipulating the customer. If you watched all their DLSS presentation, youll know. Also the Point in Nvidia comparing Port Royal's Blurry TAA with Crisp DLSS is to show that DLSS is better and it doesnt cause Blurry ness. So Nvidia screwing around with us in this case.

It is AA method by definition. And no it's not just upscaling, you would not need tensor cores to do just that.

Exactly, people are forgetting here that Tensor cores actually exist, physically on the Chip. We have plenty of Upscaling methods that are way better than current DLSS if this were the case.
 
So Nvidia screwing around with us in this case.
Of Course they are! They are stretching the truth to sell a product. AMD has done that before. Just about any marketing campaign for anything. If you accept that then life doesn't spark as much outrage. Everyone is trying to get us to spend our discretionary income on their product.
 
Last edited:
Of Course they Aare! They are stretching the truth to sell a product. AMD has done that before. Just about any marketing campaign for anything. If you accept that then life doesn't spark as much outrage. Everyone is trying to get us to spend our discretionary income on their product.

I hope they get sued for this then. In theory AI should be able to do what DLSS is, but it its still a theory then Nvidia advertised it as if they already have it and all we need to do is use RTX cards. Atleast Raytracing is super good at this point. I was actually more keep for DLSS so I could play at higher frames at 4k. But behold, you cant even turn on DLSS without DXR.

The Lies about DLSS

1. It improves performance (now we know this is actually a vague statement because it has a lot of limitations )
2. maintains or sometimes ever increases Image quality (This is also a lie because it doesnt and it actually make it look worse than current scalar methods because of the obnoxious blur effect)

Jensen also when on sayin that in some cases DLSS can give you an image that looks much much better than the resolution it is upscaling to . I think this was DLSSX2 ? So he basically said in one of the presentations that a 1440p image upscaled to 4k will look super good than the 4k image itself..
 
Jeezus! Those TAA screens don't just look like CRAP, they look utter garbage blur fest!! Even FXAA or MLAA looks much sharper than that junk! Definetely looks like fabricated screens for marketing purpose imo.
Where are the SMAA or TSSAA8x comparisson shots?? I thing AMD should invest heavily in those 2 since they are almost performace gap free.
 
Jeezus! Those TAA screens don't just look like CRAP, they look utter garbage blur fest!! Even FXAA or MLAA looks much sharper than that junk! Definetely looks like fabricated screens for marketing purpose imo.
Where are the SMAA or TSSAA8x comparisson shots?? I thing AMD should invest heavily in those 2 since they are almost performace gap free.

Agreed, dont understand why amd is betting on that taa crap.
 
taa is crap,full of motion artifacts and image smearing.
 
Not voting as none of the points really apply to me; I'm keen to see how it works firsthand, but I'm not jumping on the Turing bandwagon anytime soon...
 
taa is very blurred, i do prefer dlss on those images above, night and day compared. What the hell amd is smoking saying taa is better than dlss.

Yes on those images, the difference is very obvious. But check out real game samples, BF5 and Metro examples here on TPU.
It's as if the labels would be mixed up here, the DLSS is heavily blurred in reality.
 
Looks like AMD is scared. They were caught off guard.
 
IMO DLSS was only made to make RTX Games playable, just look at the new Metro 1440 and 4k benchmarks. DLSS will be present in future GPU generations but not mandatory.
 
this early version of dlss works best for 4k owners,that's about it. It needs high res ground image,at least 1440p.The returns are there when it comes to fps you gain but I think this case scenario is about the only one I'd use dlss for. It's far from ideal but the results are there,needs time. what was an experiment in the forst rtx series will be bettered in the second one,both in hardware and in software.so yes,amd were completely caught here.they've completely stopped innovating themselves.RVII has zero new features compared to Vega.
 
AMD Doesn't Believe in NVIDIA's DLSS, Stands for Open SMAA and TAA Solutions
In other news - water is wet.
I would be worried about their mental health if they would say Nvidia's proprietary solution is cool or something.

In fact, AMD decides to go head-on against NVIDIA's own technical presentations, comparing DLSS' image quality and performance benefits against a native-resolution, TAA-enhanced image - they say that SMAA and TAA can work equally as well without "the image artefacts caused by the upscaling and harsh sharpening of DLSS."
Nvidia left itself wide open for this one. Nice jab, though in line with the usual marketing bullshit on the topic. Native-resolution, TAA-enhanced image will always be better than DLSS image, because DLSS is not native resolution. At the same time, DLSS will be in range of 40% faster.

Taking the example of "4K" image and comparisons, test so far tend to show that DLSS image upscaled from 1440p is roughly on par with 1800p + TAA in both image quality and performance. Native 4K + TAA will be considerably slower. 1440p + TAA will be faster but uglier.

Of course, AMD may only be speaking from the point of view of a competitor that has no competing solution. however, company representatives said that they could, in theory develop something along the lines of DLSS via a GPGPU framework - a task for which AMD's architectures are usually extremely well-suited.
Correct. But they neglect to mention this approach has a clear downside of using the same compute resources that are otherwise directly used for rendering. It is not clear if and how much of CUDA cores DLSS uses in addition to Tensor cores but the impression so far is that DLSS is leaner on compute than it would be without Tensor cores.
 
Last edited:
Pointless statement from AMD, its clear as day their focus is elsewhere and this is a VERY cheap answer to somehow create the impression they're still in the game. They aren't doing jack shit with AA at this time, or with DLSS, and this is a fancy way of saying it.

That said, they aren't wrong... DLSS is dead in the water.

Metro SS's From Reddit: - user claims he was using the latest patch.


Dont even have to label these because its very obvious, one of the images is very blurry as if like there is some sort of Depth of Field effect being applied to the scene.......


:banghead::banghead:

LOL. Where is the blur? Its too dark to see... :D
 
Last edited:
Back
Top