Nvidia benchmark: GPU Benchmarks Hierarchy 2022 — Graphics Card Rankings

What Is Nvidia DLSS? A Basic Definition

DLSS 2.0 off vs DLSS 2.0 on
(Image credit: Nvidia)

DLSS stands for deep learning super sampling. It’s a type of video rendering technique that looks to boost framerates by rendering frames at a lower resolution than displayed and using deep learning, a type of AI, to upscale the frames so that they look as sharp as expected at the native resolution. For example, with DLSS, a game’s frames could be rendered at 1080p resolution, making higher framerates more attainable, then upscaled and output at 4K resolution, bringing sharper image quality over 1080p. 

This is an alternative to other rendering techniques — like temporal anti-aliasing (TAA), a post-processing algorithm — that requires an RTX graphics card and game support (see the DLSS Games section below). Games that run at lower frame rates or higher resolutions benefit the most from DLSS.  

Nvidia’s chart shows RTX 3080 performance at 4K with max graphics settings and DLSS 2. 0 Performance Mode and ray tracing on. (Image credit: Nvidia)

According to Nvidia , DLSS 2.0, the most common version, can boost framerates by 200-300% (see the DLSS 2.0 section below for more). The original DLSS is in far fewer games and we’ve found it to be less effective, but Nvidia says it can boost framerates «by over 70%.» DLSS can really come in handy, even with the best graphics cards, when gaming at a high resolution or with ray tracing, both of which can cause framerates to drop substantially compared to 1080p.

In our experience, it’s difficult to spot the difference between a game rendered at native 4K and one rendered in 1080p and upscaled to 4K via DLSS 2.0 (that’s the ‘performance’ mode with 4x upscaling). In motion, it’s almost impossible to tell the difference between DLSS 2.0 in quality mode (i.e., 1440p upscaled to 4K), though the performance gains aren’t as great.

For a comparison on how DLSS impacts game performance with ray tracing, see: AMD vs Nvidia: Which GPUs Are Best for Ray Tracing?. In that testing we only used DLSS 2.0 in quality mode (2x upscaling), and the gains are still quite large in the more demanding games.

When DLSS was first released, Nvidia claimed it showed more temporal stability and image clarity than TAA. While that might be technically true, it varies depending on the game, and we much prefer DLSS 2.0 over DLSS 1.0. An Nvidia rep confirmed to us that because DLSS requires a fixed amount of GPU time per frame to run the deep learning neural network, games running at high framerates or low resolutions may not have seen a performance boost with DLSS 1.0. 

Below is a video from Nvidia (so take it with a grain of salt), comparing Cyberpunk 2007 gameplay at both 1440p resolution and 4K with DLSS 2.0 on versus DLSS 2.0 off. 

DLSS is only available with RTX graphics cards, but AMD today released its own version of the technology, AMD Fidelity FX Super Resolution (AMD FSR). FSR is GPU agnostic, meaning it will work on Nvidia and even Intel GPUs that have the necessary hardware features. At least 10 game studios will claim FSR adoption among their games and engines this year. FSR is also available on the PlayStation 5 and will be available on the Xbox Series X and S.

DLSS Games 

In order to use DLSS, you need an RTX graphics card and need to be playing a game or other type of application that supports the feature. You can find a full list of games announced to have DLSS as by the end of July via Nvidia and below.

  • Amid Evil
  • Anthem
  • Aron’s Adventure 
  • Battlefield V
  • Bright Memory
  • Call of Duty: Black Ops Cold War
  • Call of Duty: Modern Warfare
  • Call of Duty: Warzone
  • Chernobylite
  • Control
  • CRSED: F.O.A.D. (Formerly Cuisine Royale)
  • Crysis Remastered
  • Cyberpunk 2077
  • Death Stranding
  • Deliver Us the Moon
  • Doom Eternal
  • Dying: 1983
  • Edge of Eternity
  • Enlisted
  • Everspace 2
  • F1 2020
  • Final Fantasy XV
  • Fortnite
  • Ghostrunner
  • Gu Jian Qi Tan Online
  • Icarus
  • Into the Radius VR
  • Iron Conflict
  • Justice
  • LEGO Builder’s Journey 
  • Marvel’s Avengers
  • MechWarrior 5: Mercenaries
  • Metro Exodus
  • Metro Exodus PC Enhanced Edition
  • Minecraft With RTX For Windows 10
  • Monster Hunter: World
  • Moonlight Blade
  • Mortal Shell
  • Mount & Blade II: Bannerlord
  • Necromunda: Hired Gun
  • Nine to Five
  • Naraka: Bladepoint
  • No Man’s Sky
  • Nioh 2 — The Complete Edition
  • Outriders
  • Pumpkin Jack
  • Rainbow Six Seige
  • Ready or Not
  • Red Dead Redemption 2
  • Redout: Space Assault
  • Rust
  • Scavengers
  • Shadow of the Tomb Raider
  • Supraland
  • System Shock
  • The Ascent
  • The Fabled Woods
  • The Medium
  • The Persistence
  • War Thunder
  • Watch Dogs: Legion
  • Wolfenstein: Youngblood
  • Wrench
  • Xuan-Yuan Sword VII

Note that Unreal Engine and Unity Engine both have support for DLSS 2. 0, meaning games using those engines should be able to easily implement DLSS. Nvidia also announced that Vulkan-based Linux games will be able to support DLSS on June 23, thanks to a Linux graphics driver adding support for games using Proton. 

 There are also other types of apps besides games, such as SheenCity Mars, an architectural visualization, that use DLSS. 

DLSS 2.0 and DLSS 2.1  

(Image credit: Shutterstock)

In March 2020, Nvidia announced DLSS 2.0, an updated version of DLSS that uses a new deep learning neural network that’s supposed to be up to 2 times faster than DLSS 1.0 because it leverages RTX cards’ AI processors, called Tensor Cores, more efficiently. This faster network also allows the company to remove any restrictions on supported GPUs, settings and resolutions.

DLSS 2.0 is also supposed to offer better image quality while promising up to 2-3 times the framerate (in 4K Performance Mode) compared to the predecessor’s up to around 70% fps boost. Using DLSS 2.0’s 4K Performance Mode, Nvidia claims an RTX 2060 graphics card can run games at max settings at a playable framerate. Again, a game has to support DLSS 2.0, and you need an RTX graphics card to reap the benefits.

The original DLSS was apparently limited to about 2x upscaling (Nvidia hasn’t confirmed this directly), and many games limited how it could be used. For example, in Battlefield V, if you have an RTX 2080 Ti or faster GPU, you can only enable DLSS at 4K — not at 1080p or 1440p. That’s because the overhead of DLSS 1.0 often outweighed any potential benefit at lower resolutions and high framerates.

In September 2020, Nvidia released DLSS 2.1, which added an Ultra Performance Mode for super high-res gaming (9x upscaling), support for VR games, and dynamic resolution. The latter, an Nvidia rep told Tom’s Hardware, means that, «The input buffer can change dimensions from frame to frame while the output size remains fixed. If the rendering engine supports dynamic resolution, DLSS can be used to perform the required upscale to the display resolution. » Note that you’ll often hear people referring to both the original DLSS 2.0 and the 2.1 update as «DLSS 2.0.»  

DLSS 2.0 Selectable Modes 

One of the most notable changes between the original DLSS and the fancy DLSS 2.0 version is the introduction of selectable image quality modes: Quality, Balanced, or Performance — and Ultra Performance with 2.1. This affects the game’s rendering resolution, with improved performance but lower image quality as you go through that list.

With 2.0, Performance mode offered the biggest jump, upscaling games from 1080p to 4K. That’s 4x upscaling (2x width and 2x height). Balanced mode uses 3x upscaling, and Quality mode uses 2x upscaling. The Ultra Performance mode introduced with DLSS 2.1 uses 9x upscaling and is mostly intended for gaming at 8K resolution (7680 x 4320) with the RTX 3090. While it can technically be used at lower target resolutions, the upscaling artifacts are very noticeable, even at 4K (720p upscaled). Basically, DLSS looks better as it gets more pixels to work with, so while 720p to 1080p looks good, rendering at 1080p or higher resolutions will achieve a better end result.

How does all of that affect performance and quality compared to the original DLSS? For an idea, we can turn to Control, which originally had DLSS 1.0 and then received DLSS 2.0 support when released. (Remember, the following image comes from Nvidia, so it’d be wise to take it with a grain of salt too.) 

Control at 1080p with DLSS off (top), the DLSS 1.0 on (middle) and DLSS 2.0 Quality Mode on (bottom)   (Image credit: Nvidia)

One of the improvements DLSS 2.0 is supposed to bring is strong image quality in areas with moving objects. The updated rendering in the above fan image looks far better than the image using DLSS 1.0, which actually looked noticeably worse than having DLSS off. 

DLSS 2.0 is also supposed to provide an improvement over standard DLSS in areas of the image where details are more subtle.

Control at 1440p using the original DLSS (top) and DLSS 2.0 Quality Mode (bottom) (Image credit: Nvidia)

Nvidia promised that DLSS 2.0 would result in greater game adoption. That’s because the original DLSS required training the AI network for every new game needed DLSS support. DLSS 2.0 uses a generalized network, meaning it works across all games and is trained by using «non-game-specific content,» as per Nvidia.

For a game to support the original DLSS, the developer had to implement it, and then the AI network had to be trained specifically for that game. With DLSS 2.0, that latter step is eliminated. The game developer still has to implement DLSS 2.0, but it should take a lot less work, since it’s a general AI network. It also means updates to the DLSS engine (in the drivers) can improve quality for existing games. Unreal Engine 4 and Unreal Engine 5 have DLSS 2.0 support, and Unity will add it via its 2021.2 update this year. That makes trivial for games based on those engines to enable the feature.  

How Does DLSS Work? 

Both the original DLSS and DLSS 2.0 work with Nvidia’s NGX supercomputer for training of their respective AI networks, as well as RTX cards’ Tensor Cores, which are used for AI-based rendering. 

For a game to get DLSS 1.0 support, first Nvidia had to train the DLSS AI neural network, a type of AI network called convolutional autoencoder, with NGX. It started by showing the network thousands of screen captures from the game, each with 64x supersample anti-aliasing. Nvidia also showed the neural network images that didn’t use anti-aliasing. The network then compared the shots to learn how to «approximate the quality» of the 64x supersample anti-aliased image using lower quality source frames. The goal was higher image quality without hurting the framerate too much.

The AI network would then repeat this process, tweaking its algorithms along the way so that it could eventually come close to matching the 64x quality with the base quality images via inference. The end result was «anti-aliasing approaching the quality of [64x Super Sampled], whilst avoiding the issues associated with TAA, such as screen-wide blurring, motion-based blur, ghosting and artifacting on transparencies,» Nvidia explained in 2018. 

DLSS also uses what Nvidia calls «temporal feedback techniques» to ensure sharp detail in the game’s images and «improved stability from frame to frame.» Temporal feedback is the process of applying motion vectors, which describe the directions objects in the image are moving in across frames, to the native/higher resolution output, so the appearance of the next frame can be estimated in advance.  

DLSS 2.0   (Image credit: Nvidia)

DLSS 2.0 gets its speed boost through its updated AI network that uses Tensor Cores more efficiently, allowing for better framerates and the elimination of limitations on GPUs, settings and resolutions. Team Green also says DLSS 2.0 renders just 25-50% of the pixels (and only 11% of the pixels for DLSS 2. 1 Ultra Performance mode), and uses new temporal feedback techniques for even sharper details and better stability over the original DLSS. 

Nvidia’s NGX supercomputer still has to train the DLSS 2.0 network, which is also a convolution autoencoder. Two things go into it, as per Nvidia: «low resolution, aliased images rendered by the game engine» and «low resolution, motion vectors from the same images — also generated by the game engine.»

DLSS 2.0 uses those motion vectors for temporal feedback, which the convolution autoencoder (or DLSS 2.0 network) performs by taking «the low resolution current frame and the high resolution previous frame to determine on a pixel-by-pixel basis how to generate a higher quality current frame,» as Nvidia puts it. 

The training process for the DLSS 2.0 network also includes comparing the image output to an «ultra-high-quality» reference image rendered offline in 16K resolution (15360 x 8640). Differences between the images are sent to the AI network for learning and improvements. Nvidia’s supercomputer repeatedly runs this process, on potentially tens of thousands or even millions of reference images over time, yielding a trained AI network that can reliably produce images with satisfactory quality and resolution.

With both DLSS and DLSS 2.0, after the AI network’s training for the new game is complete, the NGX supercomputer sends the AI models to the Nvidia RTX graphics card through GeForce Game Ready drivers. From there, your GPU can use its Tensor Cores’ AI power to run the DLSS 2.0 in real-time alongside the supported game.

Because DLSS 2.0 is a general approach rather than being trained by a single game, it also means the quality of the DLSS 2.0 algorithm can improve over time without a game needing to include updates from Nvidia. The updates reside in the drivers and can impact all games that utilize DLSS 2.0.

This article is part of the Tom’s Hardware Glossary.

Further reading:

  • Best Graphics Cards
  • GPU Benchmarks and Hierarchy

Get instant access to breaking news, in-depth reviews and helpful tips.

Contact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsors

Scharon Harding has a special affinity for gaming peripherals (especially monitors), laptops and virtual reality. Previously, she covered business technology, including hardware, software, cyber security, cloud and other IT happenings, at Channelnomics, with bylines at CRN UK.

Topics

Gaming

Nvidia

Nvidia GeForce RTX 4080 Review: More Efficient, Still Expensive

Tom’s Hardware Verdict

The RTX 4080 has all the technological advancements of the Ada Lovelace architecture, with a price that’s difficult to justify. It’s too close to the 4090 to entice extreme performance enthusiasts, and soon it will have to contend with AMD’s RX 7900 series. But lots of people prefer Nvidia and want DLSS and are willing to pay the piper his dues.

Pros
  • +

    Second-fastest GPU (for now)

  • +

    Much improved efficiency

  • +

    Excellent ray tracing performance

  • +

    Packs all the Ada Lovelace enhancements

Cons
  • —

    High price without the halo performance of the 4090

  • —

    Needs DLSS 3 to truly shine in gaming performance

  • —

    AMD’s RDNA 3 could provide strong competition

  • —

    Lingering concerns surrounding the 16-pin connector

Why you can trust Tom’s Hardware
Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.

The Nvidia GeForce RTX 4080 is the follow-up to last month’s RTX 4090 launch, now one of the best graphics cards and the top listing in our GPU benchmarks hierarchy. Of course, a bit of the shine has come off thanks to the melting 16-pin connectors. The good news: RTX 4080 uses less power, which should mean it’s also less likely to funnel enough power to melt the plastic connector… maybe. The bad news: At $1,199, it’s still priced out of reach for most gamers and represents a big jump in generational pricing, inheriting the RTX 3080 Ti launch price that we also felt was too high.

We already know most of what to expect from Nvidia’s Ada Lovelace architecture, so the only real question now is how performance scales down to fewer GPU shaders, less memory, less cache, a narrower memory interface, etc. Let’s quickly look at the specifications for a few of the top Nvidia and AMD GPUs. 

Swipe to scroll horizontally

Nvidia and AMD GPU Specifications
Graphics Card RTX 4080 RTX 4090 RTX 3090 Ti RTX 3080 Ti RTX 3080 RX 7900 XTX RX 7900 XT
Architecture AD103 AD102 GA102 GA102 GA102 Navi 31 Navi 31
Process Technology TSMC 4N TSMC 4N Samsung 8N Samsung 8N Samsung 8N TSMC N5 + N6 TSMC N5 + N6
Transistors (Billion) 45. 2) 378.6 608.4 628.4 628.4 628.4 300 + 222 300 + 185
SMs 76 128 84 80 68 96 84
GPU Shaders 9728 16384 10752 10240 8704 12288 10752
Tensor Cores 304 512 336 320 272 N/A N/A
Ray Tracing «Cores» 76 128 84 80 68 96 84
Boost Clock (MHz) 2505 2520 1860 1665 1710 2500 2400
VRAM Speed (Gbps) 22. 4 21 21 19 19 20 20
VRAM (GB) 16 24 24 12 10 24 20
VRAM Bus Width 256 384 384 384 320 384 320
L2 Cache 64 72 6 6 5 96 80
ROPs 112 176 112 112 96 192 192
TMUs 304 512 336 320 272 384 336
TFLOPS FP32 48. 7 82.6 40 34.1 29.8 61.4 51.6
TFLOPS FP16 (FP8/INT8) 390 (780) 661 (1321) 160 (320) 136 (273) 119 (238) 123 (246) 103 (206)
Bandwidth (GBps) 717 1008 1008 912 760 960 800
TBP (watts) 320 450 450 350 320 355 300
Launch Date Nov 2022 Oct 2022 Mar 2022 Jun 2021 Sep 2020 Dec 2022 Dec 2022
Launch Price $1,199 $1,599 $1,999 $1,199 $699 $999 $899

There’s a relatively large gap between the RTX 4080 and the larger RTX 4090. You get most of an AD103 GPU — 76 of the potential 80 Streaming Multiprocessors (SMs) — but that’s still 40% fewer GPU shaders and other functional units than the RTX 4090. Clock speeds are similar, you get 33% fewer memory channels, VRAM, and bandwidth, and the rated TBP drops by 29%. On paper, the RTX 4090 could be up to 70% faster based on the theoretical compute performance, and that’s a concern.

$1,199 is hardly affordable, so it feels like anyone even looking at the RTX 4080 should probably just save up the additional $400 for the RTX 4090 and go for broke — or melted. But then the RTX 4090 has been sold out at anywhere below $2,100 since launch, which means it could actually be a $900 upsell, and that’s far more significant.

The pricing becomes even more of a concern when we factor in AMD’s Radeon RX 7900 XTX/XT cards coming next month. We now have all the pertinent details for the first cards using AMD’s RDNA 3 GPU architecture, and they certainly look promising. Prices are still high, but the specs comparisons suggest AMD might be able to beat the RTX 4080 while costing at least $200–$300 less. This means, unless you absolutely refuse to consider purchasing an AMD graphics card, you should at least wait until next month to see what the red team has to offer.

Image 1 of 2

The RTX 4080 has seven Graphics Processing Clusters, 76 Streaming Multiprocessors, and 64MB of L2 cache. (Image credit: Nvidia)The full AD103 adds four more SMs to the bottom-left GPC. Also note that there are four NVDEC blocks and the 4080 only enables one of them. (Image credit: Nvidia)

However, Nvidia does have some extras that AMD is unlikely to match in the near term. For example, the Deep learning and AI horsepower in RTX 4080 far surpass what AMD intends to offer. If we’ve got the figures right, AMD’s FP16 and INT8 throughput will be less than a third of the RTX 4080.

Nvidia also offers DLSS 3 courtesy of the enhanced Optical Flow Accelerator (OFA). Ten games already support the technology: Bright Memory: Infinite, Destroy All Humans! 2 — Reprobed, F. I.S.T.: Forged in Shadow Torch, F1 22, Justice, Loopmancer, Marvel’s Spider-Man Remastered, Microsoft Flight Simulator, A Plague Tale: Requiem, and Super People. That’s about half as many DLSS 3 games in less than a month as those with AMD’s FSR2 technology. Of course, you need an RTX 40-series GPU for DLSS 3, while FSR2 works with pretty much everything.

Nvidia GPUs also tend to be heavily favored by professional users, or at least their employers. So while true workstations will likely opt for the RTX 6000 48GB card as opposed to a GeForce RTX 40-series, there’s certainly potential in picking up one or more RTX 4080 cards for AI and deep learning use. Content creators may also find something to like, though again, if you’re willing to pay for a 4080, it may not be a huge step up in pricing to nab a 4090 instead.

Another piece of good news (depending on which side of the aisle you fall, we suppose) is that GPU mining remains unprofitable. Gamers won’t be able to offset the price of a new graphics card through cryptocurrency mining, but at least there should be more GPUs available for gamers. Now let’s see exactly what Nvidia has to offer with its new RTX 4080. 

  • MORE: Best Graphics Cards
  • MORE: GPU Benchmarks and Hierarchy
  • MORE: All Graphics Content
  • 1

Current page:
Nvidia GeForce RTX 4080 Founders Edition Review

Next Page Nvidia RTX 4080 Design and Assembly

Jarred Walton is a senior editor at Tom’s Hardware focusing on everything GPU. He has been working as a tech journalist since 2004, writing for AnandTech, Maximum PC, and PC Gamer. From the first S3 Virge ‘3D decelerators’ to today’s GPUs, Jarred keeps up with all the latest graphics trends and is the one to ask about game performance.

Tests of video cards and games, reviews, news from the world of games and IT technologies.

Posted by svl, . Posted in Motherboards

In our review, we present to you the ASUS ROG Strix B650-A Gaming WiFi Motherboard, courtesy of ASUS.

Posted by svl, . Published in Action / FPS / TPS

In this test we present year-end review for The Callisto Protocol, which was tested at maximum ray-traced graphics quality settings.

Posted by svl, . Published in Action / FPS / TPS

In this test, we present the final annual review for Marvel’s Spider-Man: Miles Morales.

Posted by svl, . Published in Action / FPS / TPS
nine0005

In this test, we present the final annual review for the game UNCHARTED: Legacy of Thieves Collection.

Posted by svl, . Published in Action / FPS / TPS

In this test, we present the final annual review for Call of Duty: Warzone 2.0, which was tested at maximum graphics quality settings.

nine0002
Posted by svl, . Posted in Test Video Cards

Upon completion of our testing, we present our debut material to determine the fastest solutions of the past 2022. 11 of the most popular and grafonic games chosen by us were taken as a basis.

Posted by svl, . Published in RTS / Strategy

Company of Heroes 3 is a continuation of the famous RTS based on the events of World War II. This time the place of battle is the Italian Mediterranean. nine0005

Posted by zhenek, . Published in Action / FPS / TPS

In this review, we present The Witcher 3: Wild Hunt: v. 1.32 vs. v.4.0, tested on the GEFORCE GTX and RADEON RX families of video cards.

Posted by zhenek, . Published in Action / FPS / TPS

In this review, we present a test of the game Dying Light 2: Stay Human, tested on 9 families of video cards0039 GEFORCE GTX and RADEON RX , at graphics quality settings high .

Posted by zhenek, . Published in Action / FPS / TPS

In this review, we present a test of the game God of War, tested on the families of video cards GEFORCE GTX and RADEON RX , with graphics quality settings of maximum .

Nvidia 9 graphics card tester0001

System

Views: 11915

Greetings to all readers and subscribers of 3support. ru! In this article, we will talk about three free programs for checking the Nvidia video card, provide download links and teach you how to use them. With these tools, you can easily test your graphics card for performance, as well as get information about the temperature and performance of the GPU. So let’s go! nine0005

  1. FurMark
  2. Nvidia Inspector

FurMark

FurMark is a free graphics card performance test program that allows you to determine whether the installed video adapter is working without any special knowledge. The program has a simple interface, high functionality and relatively low weight. Download, install and run the application.

http://geeks-3d.ru

FurMark’s main menu looks like this:

The following utilities are integrated into this application — GPU-Z, GPU Shark and GPU burner. For testing, we need GPU-Z . With its help, we will monitor the status of the video adapter during testing. We click on the button of the same name to launch GPU-Z:

After which we will see the following:
We are interested in the “Sensors” tab:
In this tab, we can monitor the state of the video card during testing. Now we return to Furmark and run the test itself. To do this, click on the «GPU stress test» button:
Next, the program will warn us that in order to prevent unpleasant consequences (failure of the video card), our computer must meet the following requirements:

  • The graphics adapter must have good cooling
  • The system unit must have a powerful power supply
  • The CPU and GPU must run at the frequency set by the manufacturer (no overclocking).

If your system meets all these requirements, then start the test by clicking the «Go» button. Then you will see the following:
This is the test of the video card for performance. This test should be carried out 15-20 minutes . During testing, monitor the temperature in the GPU-Z application. If the temperature during all this time has not exceeded the threshold of 80-85 °C , then the video card is in order. If the temperature approached 90 ° C Celsius or even exceeded this mark, then we can say with confidence that the graphics adapter has problems. Also, during the scan, the computer may freeze or even restart. If this happens, then this means that the video card has a big malfunction. nine0005

Nvidia Inspector

Nvidia Inspector is a small free program created specifically for the Nvidia family of graphics cards. With its help, you can get the most necessary data about the video card in a matter of seconds, for example:

  • Memory and chip characteristics
  • BIOS version information
  • Installed driver details
  • Frequency
  • Fan speed information
  • GPU and memory usage information

You can also use this program to check the temperature of any Nvidia video card.

On By alexxlab