Monday, February 18th 2019

NVIDIA: Image Quality for DLSS in Metro Exodus to Be Improved in Further Updates, and the Nature of the Beast

NVIDIA, in a blog post/Q&A on its DLSS technology, promised implementation and image quality improvements on its Metro Exodus rendition of the technology. If you'll remember, AMD recently vouched for other, non-proprietary ways of achieving desired quality of AA technology across resolutions such as TAA and SMAA, saying that DLSS introduces "(...) image artefacts caused by the upscaling and harsh sharpening." NVIDIA in its blog post has dissected DLSS in its implementation, also clarifying some lingering questions on the technology and its resolution limitations that some us here at TPU had already wondered about.

The blog post describes some of the limitations in DLSS technology, and why exactly image quality issues might be popping out here and there in titles. As we knew from NVIDIA's initial RTX press briefing, DLSS basically works on top of an NVIDIA neural network. Titled the NGX, it processes millions of frames from a single game at varying resolutions, with DLSS, and compares it to a given "ground truth image" - the highest quality possible output sans any shenanigans, generated from just pure raw processing power. The objective is to train the network towards generating this image without the performance cost. This DLSS model is then made available for NVIDIA's client to download and to be run at your local RTX graphics card level, which is why DLSS image quality can be improved with time. And it also helps explain why closed implementations of the technology, such as 3D Mark's Port Royal benchmark, show such incredible image quality scenarios compared to, say, Metro Exodus - there is a very, very limited number of frames that the neural network needs to process towards achieving the best image quality.
Forumites: This is an Editorial

The nature of DLSS means that the network needs to be trained for every conceivable resolution, since different rendering resolutions will require different processing for the image to resemble the ground truth we're looking for. This is the reason for Metro Exodus' limits in DLSS - it's likely that NVIDIA didn't actually choose not to enable it at 1080p with RTX off, it was just a case of the rendering time on its NGX cluster not being enough, in time for launch, to cover all of the most popular resolutions, with or without RTX, across its three available settings. So NVIDIA mated both settings, to allow the greatest image quality and performance improvements for those gamers that want to use RTX effects, and didn't train the network for non-RTX scenarios.
This brings with it a whole lot of questions - how long exactly does NVIDIA's neural network take to train an entire game's worth of DLSS integration? With linear titles, this is likely a great technology - but apply this to an open-world setting (oh hey, like Metro Exodus) and this seems like an incredibly daunting task. NVIDIA had this to say on its blog post:
For Metro Exodus, we've got an update coming that improves DLSS sharpness and overall image quality across all resolutions that didn't make it into day of launch. We're also training DLSS on a larger cross section of the game, and once these updates are ready you will see another increase in quality. Lastly, we are looking into a few other reported issues, such as with HDR, and will update as soon as we have fixes.
So this not only speaks to NVIDIA recognizing that DLSS image quality isn't at the level it's supposed to be (which implies it can actually degrade image quality, giving further credence to AMD's remarks on the matter), but also confirms that they're constantly working on improving DLSS' performance and image quality - and more interestingly, that this is something they can always change, server-side. I'd question the sustainability of DLSS' usage, though; the number of DLSS-enabled games is low enough as it is - and yet NVIDIA seems to be having difficuly in keeping up even when it comes to AAA releases. Imagine if DLSS picked up like NVIDIA would like it (or would they?) and expand to most of the launched games. Looking at what we know, I don't even think that scenario of support would be possible - NVIDIA's neural network would be bottlenecked with all the processing time required for these games, their different rendering resolutions and RTX settings.
DLSS really is a very interesting technology that empowers the RTX graphics card of every user with the power of the cloud, as NVIDIA said it did. However, there are certainly some quirks that require more processing time than they've been given, and there are limits to how much processing power NVIDIA can and will dedicate to each title. That the network needs to be trained again and again and again for every new title out there bodes well for a controlled, NVIDIA-fed games development environment, but that's not the real world - especially not with an AMD-led console market. I'd question DLSS' longevity and success on these factors alone, whilst praising its technology and forward-thinking design immensely.
Source: NVIDIA
Add your own comment

112 Comments on NVIDIA: Image Quality for DLSS in Metro Exodus to Be Improved in Further Updates, and the Nature of the Beast

#101
Xzibit
lexluthermiesterHas the thought occurred to you that both could be true?
If you would kindly link to your proof.
Posted on Reply
#103
Crackong
AquinusOne word. Latency. We already can see the latency hit it takes when using the tensor cores (hence why the GPU needs to be under heavy load, otherwise the framerate will be too high to make it worth it.) Offloading this stuff to the CPU will make the latency problem worse, at least that's my take on it.
Yes and it is disappointed Nvidia went with the downsampling route of DLSS and hid the fact that there is a virtual framerate cap because of latency issues.
Would be nice if they kept their promises and release the "upsampling to 8k" route of DLSS, but nobody mentions it now so I guess it vanished into thin air already.
Posted on Reply
#106
Aquinus
Resident Wat-man
CrackongYes and it is disappointed Nvidia went with the downsampling route of DLSS and hid the fact that there is a virtual framerate cap because of latency issues.
Would be nice if they kept their promises and release the "upsampling to 8k" route of DLSS, but nobody mentions it now so I guess it vanished into thin air already.
It's not disappointing as much as it's predictable. I'm actually happy that nVidia is finding a use for this stuff. In this day and age with the amount of compute power on GPUs, it's insane to think that we can't do these kinds of things. There actually is quite a bit of value in doing something like DLSS. The problem is one of what you're striving for. If you goal is performance, DLSS actually isn't bad if native 4k isn't performant enough. If your goal is image quality, this is most definitely not a tool you want in your arsenal.
Posted on Reply
#107
Steevo
lexluthermiesterHave you actually seen the effect in question? 80-90fps is not "Shitty". BF5 is not that great a game, but it does show what can be done, to great effect, with raytracing. Maybe you're not seeing the subtleties?

But that illustrates perfectly the point I was making, AF took time to perfect. That doesn't mean early implementations were perfect, but they were good enough to show the benefit on offer.

That is not logical as stated.

Except that 2080ti's are still selling out and the RTX series is one of the fastest selling lines of GPU's in history. And "Physx" functionality is literally built in to every GPU NVidia makes making it one of the most adopted feature sets in the world.
Isn't this thread about DLSS and how it's implementation lowers visual quality with performance equal to a 70% render that has better visual quality.

It's like saying a car is as fast as a fighter jet, but then failing to disclose it's only while both are turned off.

What part of DLSS has worse visual quality than no DLSS is hard to understand? What part of rendering at 70% provides the same performance as DLSS is hard to understand? What part of rendering at 70% with the same performance at DLSS with better visual quality do you not understand?

Are you so blind to a brand that you really cannot see a fault here? A feature that reduces quality to regain performance that is inferior to existing technology or technique is not better, it's worse.
Posted on Reply
#108
EarthDog
SteevoIsn't this thread about DLSS and how it's implementation lowers visual quality with performance equal to a 70% render that has better visual quality.
Well, the thread title is about how they were improved in Metro...not about how bad the implementation is.

Owners need a thread just avoid this bullshit. Lol
Posted on Reply
#109
Aquinus
Resident Wat-man
EarthDogWell, the thread title is about how they were improved in Metro...not about how bad the implementation is.
A frosted dog turd is still a dog turd.
Posted on Reply
#110
EarthDog
4-4!!!

This thread IS about the frosting. :)
Posted on Reply
#111
Aquinus
Resident Wat-man
EarthDogThis thread IS about the frosting. :)
Right, when it should be about the dog turd or at the very least marketed like a dog turd. The frosting is literally the distraction. The point isn't that it's improving, the point is that it's always going to be mediocre despite improvements because it's just fancy upscaling, nothing more, nothing less. This article is more like a, "we improved image quality and it's a work in progress," when this feature was never about image quality and completely about performance. It's like saying that running at 720p instead of 480p is an improvement on a 1080p native display. Sure, it's an improvement, but it still sucks when it comes to image quality. If nVidia marketed this as what it actually was, fancy upscaling, we wouldn't have a problem.
EarthDog4-4!!!
I feel like you're trying to bait me by doing this.
Posted on Reply
#112
EarthDog
I cant help how you feel, A. Sorry.
Posted on Reply
Add your own comment
Nov 19th, 2026 20:56 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts