Nvidia DLSS upscaling is now available in over 200 games

[nanoguy]

In brief: Just as AMD is celebrating a year of availability for its FSR upscaling tech, Nvidia's DLSS is close to turning four. Developers haven't been too quick to adopt either technology, but Team Green claims bragging rights for being the first to reach the 200 game milestone.

Last year, AMD introduced its FidelityFX Super Resolution (FSR) technology — the much-awaited answer to Nvidia's Deep Learning Super Sampling (DLSS). Since then, several developers have implemented FSR into their titles, and Team Red claims over 110 existing and upcoming games support its upscaling tech -- that's between FSR and FSR 2.0 though, the former of which is generally inferior.

The company achieved this level of support relatively quickly. However, DLSS is still superior when considering the number of released games and applications that can leverage it. As of writing, Nvidia claims there are no less than 200 titles that can leverage DLSS for a healthy performance boost.

Nvidia's DLSS will also be available in upcoming titles like Warhammer 40,000: Darktide, Steelrising, and A Plague Tale: Requiem. This month, Loopmancer, Hell Pie, and others will also come with support for DLSS, allowing cards like the RTX 3060 and RTX 3070 to achieve high frame rates even at 4K with ray-traced reflections and ray-traced ambient occlusion enabled.

PCGamingWiki has a neat list of all the games that support FSR, DLSS, or both. It's worth noting there are still over 40 games that were announced to support DLSS but have yet to receive it, including the following:

• A Plague Tale: Requiem
• ARK: Survival Evolved
• Atomic Heart
• Black Myth: Wukong
• Boundary
• Darksiders 3
• Dauntless
• Dying: 1983
• F1 22
• Fear the Wolves
• The Forge Arena
• Fractured Lands
• GRIT
• Hell Pie
• Islands of Nyne: Battle Royale
• Justice Online
• JX3 Online
• Kinetik
• LEAP (available now in early access)
• Loopmancer
• Microsoft Flight Simulator 2020
• Midnight Ghost Hunt (available now in early access)
• Playerunknown's Battlegrounds
• Ratten Reich
• Remnant: From the Ashes
• Ratten Reich
• SCP: Pandemic (available now in early access)
• Serious Sam 4
• Stellrising (September)
• Stormdivers
• Super People
• Synced: Off-Planet
• System Shock (available now in the demo)
• The Anacrusis (available now in early access)
• The Day Before
• The Division 2
• The Lord of the Rings: Gollum (September)
• Turbo Sloths
• Vampire: The Masquerade – Bloodlines 2
• Voidtrain
• Warhammer 40,000: Darktide
• We Happy Few

One could argue that AMD has achieved a relatively faster adoption rate for FSR than Nvidia has managed with DLSS, while also being able to run on a wide range of hardware from both companies. Meanwhile, support for Intel's XeSS is expected to land in titles like Death Stranding Director's Cut, Ghostwire: Tokyo, Enlisted, and Dolmen sometime in the coming months.

Permalink to story.

https://www.techspot.com/news/95244-nvidia-dlss-upscaling-tech-now-available-over-200.html

 

[AMDGeForceRX4090]

Pretty sure FSR will overtake it in the end, being open source and working on both teams GPUs and consoles.

 

[nnguy2]

My comment "Claiming dlss is superior because it has more adoption " got deleted.

What kind of crap is that?

 

[scavengerspc]

My comment "Claiming dlss is superior because it has more adoption " got deleted.

What kind of crap is that?

There had to be more to it than that or this comment would have been vanquished also.

Did you call someone a terrible name?
Tell them their brain has a bad fungus?
Comment on their Mom?

Well anyway back On T.

FSR 2.0 is really impressive. I honestly can't say which is better as far as graphics quality. I can say that the performance is a little better with FSR, though my experience is limited to Dying Light 2 and RDR 2.

But damn those FSR 2 mods are popping up everywhere it seems.

 

[nnguy2]

There had to be more to it than that or this comment would have been vanquished also.

Did you call someone a terrible name?
Tell them their brain has a bad fungus?
Comment on their Mom?

Well anyway back On T.

FSR 2.0 is really impressive. I honestly can't say which is better as far as graphics quality. I can say that the performance is a little better with FSR, though my experience is limited to Dying Light 2 and RDR 2.

But damn those DSR 2 mods are popping up everywhere it seems.

Lol I might have added "so much for being objective"

 

[kiwigraeme]

Lol I might have added "so much for being objective"

Mods are pretty good for saying why for me - normally off topic in my case

 

[nnguy2]

Mods are pretty good for saying why for me - normally off topic in my case

Right. Usually I get explanation. It not this time

 

[DSirius]

Article subtitle is:
"FSR remains behind, while XeSS is nowhere to be seen outside of demos"
Interesting "perspective" in skewing data.
How many games did Nvidia implemented DLSS in 1 year and how many AMD implemented FSR in 1 year? Let's check.
Nvidia released first version of DLSS 1.0 in February 2019
https://en.wikipedia.org/wiki/Deep_learning_super_sampling
In July 2021 there were like 61 games with DLSS
https://www.dualshockers.com/nvidia-dlss-games-list/
AMD launched FSR 1.0 in June 2021
https://www.techspot.com/review/2277-amd-fsr-analysis-benchmark/
In June 2022, so after only 1 year, there are over 100 games with FSR support.
The number of games adopting FSR in 1 year is bigger than Nvidia implemented DLSS in 2 years.
The adoption rate of FSR in games is faster, more than double than DLSS.
But hey, let's write lame subtitles, instead of writing: "FSR is catching fast".
Embarrassing.

 

[b3rdm4n]

And I keep hearing that DLSS got killed, funny, it keeps being included in games...

I'm rooting for you FSR, I really am, but you have a long road of improvement and coexistence ahead of you.

 

[Neatfeatguy]

Out of all the games I play....DLSS has been an option in a couple of them and I haven't used it once. I'm not a big "new game is out, buy now, now, now!" - so I'm usually a year or two behind game releases before I get a copy, but so far I haven't been impressed with the need for DLSS or FSR.

You have one or possibly more things happening here (I'd venture to guess a bit of all):
1) DLSS/FSR is needed because hardware just can not keep up with software
2) DLSS/FSR is needed because developers aren't coding worth a damn and help is needed from the GPU/driver side of things on to make games run better
3) GPUs just can't get all the super spenders to run their games at the "magic land of: 4k @ 120fps" and all the extra help is needed so super spenders feel vindicated.
4) RT is being pushed too early and hardware can't keep up so DLSS/FSR is used to offset the loss of performance to make GPUs look more capable than they really are

 

[Fearghast]

Pretty sure FSR will overtake it in the end, being open source and working on both teams GPUs and consoles.

I am pretty sure we can't be sure of that. Considering the amount of AMD and nvidia technology that was buried and forgotten, FSR can be pretty much one of those cases.
But considering the abysmal state of 4K gaming, hardware availability, hardware capabilities, we all need FSR or any other upscaling solution to lead the way to mass implementation.

 

[rmcrys]

DLSS started first, the implementation was much better before FSR 2.0 and ATM DLSS has still a slight edge;

But let's not forget, game studios that write for consoles are already used to FSR, so games that come to console AND PC, will be easy to have FSR on PC.

On the long run I believe FSR will come first as it runs on most hardware, easy to implement on all OSes, it is open source. And after this moment, I don't expect any big improvements on the tech itself.

 

[hahahanoobs]

DLSS started first, the implementation was much better before FSR 2.0 and ATM DLSS has still a slight edge;

But let's not forget, game studios that write for consoles are already used to FSR, so games that come to console AND PC, will be easy to have FSR on PC.

On the long run I believe FSR will come first as it runs on most hardware, easy to implement on all OSes, it is open source. And after this moment, I don't expect any big improvements on the tech itself.

Let's not forget, FSR is open source now meaning AMD isn't the sole driver. They did this with Mantle and arguably Freesync.

 

[rmcrys]

Let's not forget, FSR is open source now meaning AMD isn't the sole driver. They did this with Mantle and arguably Freesync.

It's true, nevertheless let's not forget (for example) VHS vs. Betamax, where the best format doesn't always win, it comes all down to adoption. If Nvidia sells more GPUs to gamers than AMD + Intel (dGPUs) together, as well as it signs more contracts with exclusivity, it will win. I hope not...

 

[envirovore]

It's true, nevertheless let's not forget (for example) VHS vs. Betamax, where the best format doesn't always win, it comes all down to adoption. If Nvidia sells more GPUs to gamers than AMD + Intel (dGPUs) together, as well as it signs more contracts with exclusivity, it will win. I hope not...

So...what you're saying is when the porn industry picks a side on one of these technologies, that'll be the declared winner.

As the porn industry is what killed betamax off when they chose to side with the VHS format.

Edit: being facetious, of course.

 

[hahahanoobs]

It's true, nevertheless let's not forget (for example) VHS vs. Betamax, where the best format doesn't always win, it comes all down to adoption. If Nvidia sells more GPUs to gamers than AMD + Intel (dGPUs) together, as well as it signs more contracts with exclusivity, it will win. I hope not...

FSR doesn't have an experienced software dev team or the market share [AMD].

 

[b3rdm4n]

It's so strange to me that all over the internet Betamax vs VHS is one of the go to arguments for FSR eventually beating DLSS, is that really the best they can come up with?

 

[poltevo]

Article subtitle is:
"FSR remains behind, while XeSS is nowhere to be seen outside of demos"
Interesting "perspective" in skewing data.
How many games did Nvidia implemented DLSS in 1 year and how many AMD implemented FSR in 1 year? Let's check.
Nvidia released first version of DLSS 1.0 in February 2019
https://en.wikipedia.org/wiki/Deep_learning_super_sampling
In July 2021 there were like 61 games with DLSS
https://www.dualshockers.com/nvidia-dlss-games-list/
AMD launched FSR 1.0 in June 2021
https://www.techspot.com/review/2277-amd-fsr-analysis-benchmark/
In June 2022, so after only 1 year, there are over 100 games with FSR support.
The number of games adopting FSR in 1 year is bigger than Nvidia implemented DLSS in 2 years.
The adoption rate of FSR in games is faster, more than double than DLSS.
But hey, let's write lame subtitles, instead of writing: "FSR is catching fast".
Embarrassing.

Lets set aside the fact that FSR 1.0 is vastly inferior to DLSS and Techspot should only be comparing FSR 2.0 to DLSS. Your numbers don't tell the whole story and don't support your conclusion that "FSR is catching fast", even for FSR 1.0.

Using your numbers, its clear that DLSS support is accelerating, with an extra 140 games in the last year. FSR 1.0 could only manage 100 even with its ease of implementation and wider hardware support. Lots of gamers will be upgrading their GPUs soon, so the hardware support for DLSS will increase. This should speed up DLSS support further.

I've said this before: FSR 1.0 was pure marketing from AMD to justify their high prices. We saw many outlets and people praising FSR 1.0 and this contributes to high AMD prices. This is anti-consumer, since it gives the impression that AMD and Nvidia cards are feature equivalent.

FSR 2.0 is much better, even if it lags behind DLSS, but it currently lacks game support. The solution is simple: AMD cards should be priced significantly lower then equivalent Nvidia cards.

Note: I'm ignoring RT performance, but it depends how important gamers think this is.

 

[poltevo]

FSR doesn't have an experienced software dev team or the market share [AMD].

This is sadly true. There is a massive difference between AMD and Nvidia’s market share and R&D budgets. AMD can still compete on some metrics, such as raster but they need to lower prices to make up for other missing features. Personally, if RDNA3 includes a fast encoder and is a bit cheaper then an equivalent RTX card then I would be tempted.

 

[scavengerspc]

There is a massive difference between AMD and Nvidia’s market share and R&D budgets.

But market share has nothing to do with it when FSR works on everything.
And as far as R&D, it's open source, so the development is only limited by the size of the planet.

 

[DSirius]

Lets set aside the fact that FSR 1.0 is vastly inferior to DLSS and Techspot should only be comparing FSR 2.0 to DLSS. Your numbers don't tell the whole story and don't support your conclusion that "FSR is catching fast", even for FSR 1.0.

Using your numbers, its clear that DLSS support is accelerating, with an extra 140 games in the last year. FSR 1.0 could only manage 100 even with its ease of implementation and wider hardware support. Lots of gamers will be upgrading their GPUs soon, so the hardware support for DLSS will increase. This should speed up DLSS support further.

I've said this before: FSR 1.0 was pure marketing from AMD to justify their high prices. We saw many outlets and people praising FSR 1.0 and this contributes to high AMD prices. This is anti-consumer, since it gives the impression that AMD and Nvidia cards are feature equivalent.

FSR 2.0 is much better, even if it lags behind DLSS, but it currently lacks game support. The solution is simple: AMD cards should be priced significantly lower then equivalent Nvidia cards.

Note: I'm ignoring RT performance, but it depends how important gamers think this is.

DLSS 1.0 is the most inferior among upscaling technologies and in fact was a joke, and was so for more than a year - 14 months - until April 2020 Nvidia managed to release DLSS 2.0
FSR 1.0 is better than DLSS 1.0 and is working on Nvidia cards on which own Nvidia DLSS is not.
This speaks highly about Nvidia anti-user policies.
Between DLSS 1.xx (until DLSS 2.0) and FSR 1.0 the lame marketing was NVIDIA one, in fact NVIDIA mislead customers with DLSS 1.x on RTX 2xxx cards and Hardware reviewers and youtubers exposed it. But this is history.

Deep learning super sampling - Wikipedia

en.wikipedia.org

Now consumers has 2 good upscaling technologies FSR 2.x and DLSS 2.xx and adopt ration of FSR is higher because AMD made it open source.
The rest is a matter of choice and preference for buyers.

 

[b3rdm4n]

But market share has nothing to do with it when FSR works on everything.
And as far as R&D, it's open source, so the development is only limited by the size of the planet.

lets see how that works out for FSR, I hope well! but being open source is a massive 'elevator' to put it all on to advance.

FWIW, I've said it before and I said it again, I am 100% happy for an open future, and it's basically inevitable, but improvement still needs to occur, by quite some margin. Or, if openness and ease of implementation was literally all that mattered, FSR 1.0 would be in everything and DLSS would already be dead in the water, but here we are.

FSR 1.0 is better than DLSS 1.0 and is working on Nvidia cards on which own Nvidia DLSS is not.

And DLSS 1.0 precedes them all, and was attempting to do something the market didn't yet have. We have 1.0 to thank for 2.0+ and FSR of any flavour for that matter. In one way or another, it all stands on DLSS's shoulders.

 

[poltevo]

But market share has nothing to do with it when FSR works on everything.
And as far as R&D, it's open source, so the development is only limited by the size of the planet.

That's not my point. Nvidia has a large market share, makes a boat load of money and can spend a lot on R&D spend. AMDs solutions are typical for an underdog with limited R&D and market share - they are not as good and work on a wider range of hardware.

Being open source helps with wider usage, but it doesn't necessarily lead to improvements to the core algorithm. I doubt the core of FSR 2.0 (or 1.0) will improved much due to community contributions. AMD have a very good team working on both FSR versions, so any improvements will likely come from them.

Open source has proven benefits, but it doesn't replace R&D, especially when it comes to cutting edge technology.