Content deleted Content added
Davide King (talk | contribs) m ce top order, MOS:ORDER |
Citation bot (talk | contribs) Altered title. Added work. | Use this bot. Report bugs. | Suggested by 16dvnk | Category:Artificial intelligence | #UCB_Category 92/198 |
||
(47 intermediate revisions by 14 users not shown) | |||
Line 1:
{{short description|Image upscaling technology by Nvidia}}
{{Puffery|date=March 2024}}
'''Deep
== History ==
Nvidia advertised DLSS as a key feature of the [[GeForce 20 series]] cards when they launched in September 2018.<ref name="techspot">{{cite web|url=https://www.techspot.com/article/1992-nvidia-dlss-2020/|title=Nvidia DLSS in 2020: stunning results|publisher=techspot.com|date=2020-02-26|access-date=2020-04-05}}</ref> At that time, the results were limited to a few video games,
In April 2020, Nvidia advertised and shipped an improved version of DLSS named DLSS 2.0 with [[Device driver|driver]] version 445.75. DLSS 2.0 was available for a few existing games including ''Control'' and ''[[Wolfenstein: Youngblood]]'', and would later be added to many newly released games and [[game engine]]s such as [[Unreal Engine]] and [[Unity (game engine)|Unity]].<ref>{{Cite web|date=2021-02-11|title=NVIDIA DLSS Plugin and Reflex Now Available for Unreal Engine|url=https://developer.nvidia.com/blog/nvidia-dlss-and-reflex-now-available-for-unreal-engine-4-26/|access-date=2022-02-07|website=NVIDIA Developer Blog|language=en-US}}</ref>
In January 2025, Nvidia stated that there are over 540 games and apps supporting DLSS, and that over 80% of [[Nvidia RTX]] users activate DLSS.<ref>{{Cite web |date=2025-01-16 |title=DLSS enabled by over 80% of GeForce RTX gaming GPU owners, claims Nvidia |url=https://www.pcgamesn.com/nvidia/dlss-80-percent |access-date=2025-01-31 |website=PCGamesN |language=en-US}}</ref>
▲In April 2020, Nvidia advertised and shipped an improved version of DLSS named DLSS 2.0 with [[Device driver|driver]] version 445.75. DLSS 2.0 was available for a few existing games including ''Control'' and ''[[Wolfenstein: Youngblood]]'', and would later be added to many newly released games and [[game engine]]s such as [[Unreal Engine]]<ref>{{Cite web|date=2021-02-11|title=NVIDIA DLSS Plugin and Reflex Now Available for Unreal Engine|url=https://developer.nvidia.com/blog/nvidia-dlss-and-reflex-now-available-for-unreal-engine-4-26/|access-date=2022-02-07|website=NVIDIA Developer Blog|language=en-US}}</ref> and [[Unity (game engine)|Unity]].<ref>{{Cite web|date=2021-04-14|title=NVIDIA DLSS Natively Supported in Unity 2021.2|url=https://developer.nvidia.com/blog/nvidia-dlss-natively-supported-in-unity-2021-2/|access-date=2022-02-07|website=NVIDIA Developer Blog|language=en-US}}</ref> This time Nvidia said that it used the Tensor Cores again, and that the AI did not need to be trained specifically on each game.<ref name="techspot"/><ref name="gamersnexus">{{cite web|url=https://www.gamersnexus.net/news-pc/3572-hw-news-crysis-remastered-ray-tracing-on-amd-nvidia|title=HW News - Crysis Remastered Ray Tracing, NVIDIA DLSS 2, Ryzen 3100 Rumors|date=2020-04-19|access-date=2020-04-19|quote=''The original DLSS required training the AI network for each new game. DLSS 2.0 trains using non-game-specific content, delivering a generalized network that works across games. This means faster game integrations, and ultimately more DLSS games.''|archive-date=2020-09-26|archive-url=https://web.archive.org/web/20200926224142/https://www.gamersnexus.net/news-pc/3572-hw-news-crysis-remastered-ray-tracing-on-amd-nvidia|url-status=dead}}</ref> Despite sharing the DLSS branding, the two iterations of DLSS differ significantly and are not backwards-compatible.<ref name="NVIDIA">Edward Liu, NVIDIA [https://developer.nvidia.com/gtc/2020/video/s22698-vid "DLSS 2.0 - Image Reconstruction for Real-time Rendering with Deep Learning"]</ref><ref name=":1">{{Cite web|title=Truly Next-Gen: Adding Deep Learning to Games & Graphics (Presented by NVIDIA)|url=https://www.gdcvault.com/play/1026184/Truly-Next-Gen-Adding-Deep|access-date=2022-02-07|website=www.gdcvault.com}}</ref>
In March 2025, there were more than 100 games that support DLSS 4, according to Nvidia.<ref>{{Cite web |last=Mujtaba |first=Hassan |date=2025-03-13 |title=NVIDIA DLSS 4 Now In Over 100 Games With More Titles Coming Soon, Neural Shading Support For DirectX Arriving Next Month |url=https://wccftech.com/nvidia-dlss-4-now-in-over-100-games-more-titles-coming-soon-neural-shading-support-directx-next-month/ |access-date=2025-05-20 |website=Wccftech |language=en-US}}</ref> By May 2025, over 125 games supported DLSS 4.<ref>{{Cite web |last=Mujtaba |first=Hassan |date=2025-05-19 |title=NVIDIA DLSS 4 Now Available In Over 125 Games & Apps, DOOM: The Dark Ages Path Tracing Update In June & Even More DLSS Titles |url=https://wccftech.com/nvidia-dlss-4-125-games-doom-the-dark-ages-path-tracing-update-june-more-dlss-titles/ |access-date=2025-05-20 |website=Wccftech |language=en-US}}</ref><ref>{{Cite web |last=Palumbo |first=Alessio |date=2025-05-19 |title=NVIDIA DLSS 4 Multi Frame Generation and Other RTX Updates Shown Off for Upcoming and Existing PC Games |url=https://wccftech.com/nvidia-dlss-4-multi-frame-generation-and-other-rtx-updates-shown-off-for-upcoming-and-existing-pc-games/ |access-date=2025-05-20 |website=Wccftech |language=en-US}}</ref>
The first [[video game console]] to use DLSS, the [[Nintendo Switch 2]], was released on June 5, 2025.<ref>{{Cite news |last=Stuart |first=Keith |date=2025-06-05 |title=The Nintendo Switch 2 is out – here's everything you need to know |url=https://www.theguardian.com/games/2025/jun/05/the-nintendo-switch-2-is-out-today-heres-everything-you-need-to-know |access-date=2025-06-09 |work=The Guardian |language=en-GB |issn=0261-3077}}</ref>
=== Release history ===
Line 21 ⟶ 25:
|1.0||February 2019||Predominantly spatial image upscaler, required specifically trained for each game integration, included in ''[[Battlefield V]]'' and ''[[Metro Exodus]],'' among others<ref name="battlefieldv"/>
|-
|"1.9" (unofficial name)||August 2019||DLSS 1.0 adapted for running on the CUDA shader cores instead of tensor cores, used for ''[[Control (video game)|Control]]''<ref name="eurogamer"/><ref name="techspot"/><ref name="nividiacontrol">{{cite web |last1=Edelsten |first1=Andrew |title=NVIDIA DLSS: Control and Beyond |url=https://www.nvidia.com/en-us/geforce/news/dlss-control-and-beyond/ |publisher=
|-
|2.0||April 2020||An AI accelerated form of [[Temporal anti-aliasing|TAA]]U using Tensor Cores, and trained generically<ref name="control2">{{cite web|url=https://www.techquila.co.in/nvidia-dlss-2-control-review/|title=NVIDIA DLSS 2.0 Review with Control – Is This Magic?|work=TechQuila |publisher=techquila.co.in|date=2020-04-05|access-date=2020-04-06}}</ref>
|-
|3.0
Line 31 ⟶ 35:
|3.5
|September 2023
|DLSS 3.5 adds
|-
|4.0
|January 2025
|DLSS 4.0 adds Multi Frame Generation, new AI-model based on the [[Transformer (deep learning architecture)|transformer architecture]], improving frame stability, reduced memory usage, and increased lighting detail.<ref name=":6" /><ref>{{Cite web |last=Khan |first=Sarfraz |date=2025-01-14 |title=NVIDIA Confirms Updated DLSS Frame Generation On RTX 40 GPUs, Leads to Lower VRAM Usage & Faster Performance |url=https://wccftech.com/nvidia-dlss-frame-generation-rtx-40-gpus-new-ai-model-faster-lower-vram-requirement/ |access-date=2025-01-14 |website=Wccftech |language=en-US}}</ref>
|}
Line 48 ⟶ 56:
|Ultra Quality<ref name=":5">{{cite web |title=NVIDIA preparing Ultra Quality mode for DLSS, 2.2.9.0 version spotted |url=https://videocardz.com/newz/nvidia-preparing-ultra-quality-mode-for-dlss-2-2-9-0-version-spotted |access-date=2021-07-06 |website=VideoCardz.com |language=en-US}}</ref><sub> (unused)</sub>
|1.32x
|
|-
|Quality
Line 81 ⟶ 89:
The first iteration of DLSS is a predominantly spatial image upscaler with two stages, both relying on [[Convolutional neural network|convolutional]] [[Autoencoder|auto-encoder]] [[neural network]]s.<ref>{{Cite web|date=2018-09-19|title=DLSS: What Does It Mean for Game Developers?|url=https://developer.nvidia.com/blog/dlss-what-does-it-mean-for-game-developers/|access-date=2022-02-07|website=NVIDIA Developer Blog|language=en-US}}</ref> The first step is an image enhancement network which uses the current frame and motion vectors to perform [[edge enhancement]], and [[spatial anti-aliasing]]. The second stage is an image upscaling step which uses the single raw, low-resolution frame to upscale the image to the desired output resolution. Using just a single frame for upscaling means the neural network itself must generate a large amount of new information to produce the high resolution output, this can result in slight [[Hallucination (artificial intelligence)|hallucination]]s such as leaves that differ in style to the source content.<ref name="NVIDIA" />
The neural networks are trained on a per-game basis by generating a "perfect frame" using traditional [[supersampling]] to 64 samples per pixel, as well as the motion vectors for each frame. The data collected must be as comprehensive as possible, including as many levels, times of day, graphical settings, resolutions, etc. as possible. This data is also [[Data augmentation|augmented]] using common augmentations such as rotations, colour changes, and random noise to help generalize the test data. Training is performed on Nvidia's Saturn V supercomputer.<ref name=":1" /><ref name="nvidia10">{{cite web|url=https://www.nvidia.com/en-us/geforce/news/nvidia-dlss-your-questions-answered/|title=NVIDIA DLSS: Your Questions, Answered|publisher=[[Nvidia]]|date=2019-02-15|access-date=2020-04-19|quote=
This first iteration received a mixed response, with many criticizing the often soft appearance and artifacts in certain situations;<ref name="nvidia20">{{cite web|date=2020-03-23|title=NVIDIA DLSS 2.0: A Big Leap In AI Rendering|url=https://www.nvidia.com/en-us/geforce/news/nvidia-dlss-2-0-a-big-leap-in-ai-rendering/|access-date=2020-04-07|publisher=[[Nvidia]]}}</ref><ref name=":0" /><ref name="battlefieldv" /> likely a side effect of the limited data from only using a single frame input to the neural networks which could not be trained to perform optimally in all scenarios and [[Edge case|edge-cases]].<ref name="NVIDIA" /><ref name=":1" /> Nvidia also demonstrated the ability for the auto-encoder networks to learn the ability to recreate [[Depth of field|depth-of-field]] and [[motion blur]],<ref name=":1" /> although this functionality has never been included in a publicly released product.{{Citation needed|date=February 2022}}
=== DLSS 2.0 ===
DLSS 2.0 is a [[temporal anti-aliasing]] [[upsampling]] (TAAU) implementation, using data from previous frames extensively through sub-pixel jittering to resolve fine detail and reduce aliasing. The data DLSS 2.0 collects includes: the raw low-resolution input, [[motion vector]]s, [[Z-buffering|depth buffers]], and [[Exposure value|exposure]] / brightness information.<ref name="NVIDIA" /> It can also be used as a simpler TAA implementation where the image is rendered at 100% resolution, rather than being upsampled by DLSS, Nvidia brands this as [[DLAA]] (
TAA(U) is used in many modern video games and [[game engine]]s
DLSS 2.0 uses a [[Convolutional neural network|convolutional]] [[Autoencoder|auto-encoder]] [[neural network]]<ref name="nvidia20" /> trained to identify and fix temporal artifacts, instead of manually programmed heuristics as mentioned above. Because of this, DLSS 2.0 can generally resolve detail better than other TAA and TAAU implementations, while also removing most temporal artifacts. This is why DLSS 2.0 can sometimes produce a sharper image than rendering at higher, or even native resolutions using traditional TAA. However, no temporal solution is perfect, and artifacts (ghosting in particular) are still visible in some scenarios when using DLSS 2.0.
Because temporal artifacts occur in most art styles and environments in broadly the same way, the neural network that powers DLSS 2.0 does not need to be retrained when being used in different games. Despite this, Nvidia does frequently ship new minor revisions of DLSS 2.0 with new titles,<ref>{{Cite web|title=NVIDIA DLSS DLL (2.3.7) Download|url=https://www.techpowerup.com/download/nvidia-dlss-dll/|access-date=2022-02-10|website=TechPowerUp|language=en}}</ref> so this could suggest some minor training optimizations may be performed as games are released, although Nvidia does not provide changelogs for these minor revisions to confirm this. The main advancements compared to DLSS 1.0 include: Significantly improved detail retention, a generalized neural network that does not need to be re-trained per-game, and ~2x less overhead (~1–2 ms vs ~2–4 ms).<ref name="NVIDIA" />
It should also be noted that forms of TAAU such as DLSS 2.0 are not [[Video scaler|upscalers]] in the same sense as techniques such as ESRGAN or DLSS 1.0, which attempt to create new information from a low-resolution source; instead, TAAU works to recover data from previous frames, rather than creating new data. In practice, this means low resolution [[Texture mapping|textures]] in games will still appear low-resolution when using current TAAU techniques. This is why Nvidia recommends game developers use higher resolution textures than they would normally for a given rendering resolution by applying a mip-map bias when DLSS 2.0 is enabled.<ref name="NVIDIA" />
=== DLSS 3.0 ===
Augments DLSS 2.0 by making use of [[motion interpolation]]. The DLSS
=== DLSS 3.5 ===▼
DLSS 3.5 adds
=== DLSS 4.0 ===
The fourth generation of Deep Learning Super Sampling (DLSS) was unveiled alongside the [[GeForce RTX 50 series]]. DLSS 4 upscaling uses a new vision [[Transformer (deep learning architecture)|transformer]]-based model for enhanced image quality with reduced ghosting and greater image stability in motion compared to the previous [[convolutional neural network]] (CNN) model.<ref>{{cite web |last1=Leadbetter |first1=Richard |date=January 7, 2025 |title=Hands-on with DLSS 4 on Nvidia's new GeForce RTX 5080 |url=https://www.eurogamer.net/digitalfoundry-2025-hands-on-with-dlss-4-on-nvidias-new-geforce-rtx-5080 |website=Eurogamer |language=en-GB |access-date=January 7, 2025}}</ref> DLSS 4 allows a greater number of frames to be generated and [[Frame interpolation|interpolated]] based on a single traditionally rendered frame. This form of frame generation called Multi Frame Generation is exclusive to the GeForce RTX 50 series while the [[GeForce RTX 40 series]] is limited to one interpolated frame per traditionally rendered frame. According to Nvidia, this technique will increase performance by up to 800% while retaining low latency with [[Nvidia Reflex]].<ref name=":62">{{Cite web |title=NVIDIA Blackwell GeForce RTX 50 Series Opens New World of AI Computer Graphics |url=https://nvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphics |access-date=2025-01-07 |website=NVIDIA Newsroom |language=en-us}}</ref> Nvidia claims that DLSS 4x Frame Generation model uses 30% less video memory with the example of ''[[Warhammer 40,000: Darktide]]'' using 400MB less memory at 4K resolution with Frame Generation enabled.<ref>{{cite web |last1=Lin |first1=Henry |last2=Burnes |first2=Andrew |date=January 6, 2025 |title=Nvidia DLSS 4 Introduces Multi Frame Generation & Enhancements For All DLSS Technologies |url=https://www.nvidia.com/en-us/geforce/news/dlss4-multi-frame-generation-ai-innovations/ |website=Nvidia |language=en-US |access-date=January 7, 2025}}</ref> Nvidia claims that 75 games will integrate DLSS 4 Multi Frame Generation at launch, including ''[[Alan Wake 2]]'', ''[[Cyberpunk 2077]]'', ''[[Indiana Jones and the Great Circle]]'', and ''[[Star Wars Outlaws]]''.<ref>{{cite web |last1=Mujtaba |first1=Hassan |date=January 6, 2025 |title=Nvidia DLSS 4 Delivers An Insane 8x Performance Boost Versus DLSS 3 With Multi Frame Generation Technology, Enhanced Upscaling For RTX 20 & Above |url=https://wccftech.com/nvidia-dlss-4-8x-faster-dlss-3-multi-frame-generation-enhanced-upscaling-rtx-20-above/ |website=Wccftech |language=en-US |access-date=January 7, 2025}}</ref>
{| class="wikitable plainrowheaders" style="text-align:left; font-size:90%;
▲=== DLSS 3.5 ===
! style="width:12em; height:3em;" |
! [[GeForce RTX 20 series]]
! [[GeForce RTX 30 series]]
! [[GeForce RTX 40 series]]
! [[GeForce RTX 50 series]]
|-
! scope="row" style"height:2em;" | Transformer Model
| {{ya}}
| {{ya}}
| {{ya}}
| {{ya}}
|-
! scope="row" style"height:2em;" | 2× Frame Generation
▲DLSS 3.5 adds ray reconstruction, replacing multiple denoising algorithms with a single AI model trained on five times more data than DLSS 3. Ray reconstruction is available on all RTX GPUs and first targeted games with [[path tracing]] (aka "full ray tracing"), including ''[[Cyberpunk 2077]]'''s ''[[Phantom Liberty]]'' DLC, ''[[Portal with RTX]]'', and ''[[Alan Wake 2]]''.<ref name="eurogamerdlss35" /><ref name="vergedlss35" />
| {{na}}
| {{na}}
| {{ya}}
| {{ya}}
|-
! scope="row" style"height:2em;" | 3–4× Frame Generation
== Anti-aliasing ==▼
| {{na}}
| {{na}}
| {{na}}
| {{ya}}
|-
|}
=== Manually upgrading DLSS support ===
It operates on similar principles to [[Temporal anti-aliasing|TAA]]. Like TAA, it uses information from past frames to produce the current frame. Unlike TAA, DLSS does not sample every pixel in every frame. Instead, it samples different pixels in different frames and uses pixels sampled in past frames to fill in the unsampled pixels in the current frame. DLSS uses machine learning to combine samples in the current frame and past frames, and it can be thought of as an advanced and superior TAA implementation made possible by the available tensor cores.<ref name="NVIDIA"/>▼
Users can manually replace the [[Dynamic-link library|DLLs]] in games to support a newer version of DLSS. DLSS Swapper, an [[open source]] utility, can automatically do this for all installed games.<ref>{{cite web|url=https://www.pcgamer.com/hardware/this-open-source-tool-updates-dlss-to-the-latest-version-in-all-your-games-at-once-and-no-matter-the-launcher/|title=This open source tool updates DLSS to the latest version in all your games at once and no matter the launcher|last=Edser|first=Andy|work=[[PC Gamer]]|date=2024-08-30|accessdate=2025-01-28}}</ref> Replacing DLL files can not add DLSS support or features to games that do not already implement them, though some [[Video game modding|mods]] can add frame generation support.<ref>{{cite web|url=https://www.tomshardware.com/pc-components/gpus/dlss-swapper-now-updates-fsr-xess-and-dlss-too-supports-all-major-upscaling-frame-gen-technologies|title=DLSS Swapper now updates FSR, XeSS, and DLSS, too — Supports all major upscaling/frame gen technologies|last=Nasir|first=Hassam|work=[[Tom's Hardware]]|date=2025-01-27|accessdate=2025-01-28}}</ref>
▲== Anti-aliasing ==
▲DLSS requires and applies its own [[anti-aliasing]] method. Thus, depending on the game and quality setting used, using DLSS may improve image quality even over native resolution rendering.<ref>{{Cite web |last=Smith |first=Matthew S. |date=2023-12-28 |title=What Is DLSS and Why Does it Matter for Gaming? |url=https://www.ign.com/articles/what-is-nvidia-dlss-meaning |access-date=2024-06-13 |website=IGN |language=en}}</ref> It operates on similar principles to [[Temporal anti-aliasing|TAA]]. Like TAA, it uses information from past frames to produce the current frame. Unlike TAA, DLSS does not sample every pixel in every frame. Instead, it samples different pixels in different frames and uses pixels sampled in past frames to fill in the unsampled pixels in the current frame. DLSS uses machine learning to combine samples in the current frame and past frames, and it can be thought of as an advanced and superior TAA implementation made possible by the available tensor cores.<ref name="NVIDIA" /> [[Nvidia]] also offers [[Deep Learning Anti-Aliasing]] (DLAA), which provides the same AI-driven anti-aliasing DLSS uses, but without any upscaling or downscaling.<ref name=":4" />
== Architecture ==
With the exception of the shader-core version implemented in ''Control'', DLSS is only available on [[GeForce 20 series|GeForce RTX 20]], [[GeForce 30 series|GeForce RTX 30]], [[GeForce 40 series|GeForce RTX 40]], [[GeForce 50 series|GeForce RTX 50]], and [[Quadro#Quadro RTX|Quadro RTX]] series of video cards, using dedicated [[AI accelerator]]s called '''Tensor Cores'''.<ref name="nvidia20"/>{{Failed verification|date=March 2024}} Tensor Cores are available since the Nvidia [[Volta (microarchitecture)|Volta]] [[graphics processing unit|GPU]] [[microarchitecture]], which was first used on the [[Nvidia Tesla|Tesla V100]] line of products.<ref>
{{cite web|url=https://www.tomshardware.com/news/nvidia-tensor-core-tesla-v100,34384.html|title=On Tensors, Tensorflow, And Nvidia's Latest 'Tensor Cores'|publisher=tomshardware.com|date=2017-04-11|access-date=2020-04-08}}</ref> They are used for doing [[Multiply–accumulate operation|fused multiply-add]] (FMA) operations that are used extensively in neural network calculations for applying a large series of multiplications on weights, followed by the addition of a bias. Tensor cores can operate on FP16, INT8, INT4, and INT1 data types. Each core can do 1024 bits of FMA operations per clock, so 1024 INT1, 256 INT4, 128 INT8, and 64 FP16 operations per clock per tensor core, and most Turing GPUs have a few hundred tensor cores.<ref>{{Cite web|title=
==
The transformer-based AI upscaling model introduced with DLSS 4 received moderate praise for its improved image quality with regard to increased stability, reduced ghosting, better anti-aliasing, and higher level of detail, as well as its backward compatability and higher training scalability regarding future improvements.<ref>{{Cite news |title=NVIDIA DLSS 4 Transformer Review - Better Image Quality for Everyone |url=https://www.techpowerup.com/review/nvidia-dlss-4-transformers-image-quality/ |archive-url=http://web.archive.org/web/20250128024629/https://www.techpowerup.com/review/nvidia-dlss-4-transformers-image-quality/ |archive-date=2025-01-28 |access-date=2025-01-31 |work=TechPowerUp |language=en}}</ref><ref>{{Cite web |last=Leadbetter |first=Richard |date=2025-01-07 |title=Hands-on with DLSS 4 on Nvidia's new GeForce RTX 5080 |url=https://www.eurogamer.net/digitalfoundry-2025-hands-on-with-dlss-4-on-nvidias-new-geforce-rtx-5080 |access-date=2025-01-31 |website=Eurogamer.net |language=en}}</ref>
== See also ==
* [[GPUOpen#FidelityFX Super Resolution|FidelityFX Super Resolution]] – competing technology from [[AMD]]
* [[Intel XeSS]] – competing technology from [[Intel]]
* [[PlayStation Spectral Super Resolution]] – similar technology from [[
== References ==
Line 143 ⟶ 169:
== External links ==
* {{official website}}
* [https://developer.nvidia.com/rtx/dlss DLSS on
{{NVIDIA}}
Line 150 ⟶ 176:
[[Category:Nvidia]]
[[Category:Anti-aliasing algorithms]]
[[Category:Artificial intelligence]]
|