Nvidia DLAA An Anti-Aliasing Explainer | ENBLE

Nvidia DLAA An Anti-Aliasing Explainer | ENBLE

Nvidia DLAA: Enhancing Image Quality in Gaming

Nvidia DLAA

After nearly two years of development, Nvidia’s Deep Learning Anti-Aliasing (DLAA) has made its way into a long list of games, including popular titles such as Diablo IV, Baldur’s Gate 3, and Marvel’s Spider-Man. DLAA is an exclusive feature for Nvidia’s RTX graphics cards and aims to improve the final image quality in games. Unlike Nvidia’s other widely used feature, Deep Learning Super Sampling (DLSS), which focuses on performance optimization, DLAA is all about enhancing visual quality.

Understanding Nvidia DLAA

Anti-aliasing is a technique used to solve the problem of “jaggies” or the blocky, stair-stepping effect that appears on diagonal lines in video games. Nvidia DLAA is an anti-aliasing feature that leverages the same technology as DLSS. However, instead of upscaling the image, DLAA focuses on improving anti-aliasing at the native resolution.

There are three main anti-aliasing techniques: multi-sampling anti-aliasing (MSAA), fast approximate anti-aliasing (FXAA), and temporal anti-aliasing (TAA). These techniques aim to fill in the gaps between pixels and create a smoother edge on objects. DLAA works similarly to TAA, but with the added advantage of machine learning. DLAA only samples pixels that have changed from one frame to the next, providing more information for anti-aliasing, resulting in a cleaner image.

How does Nvidia DLAA work?

DLAA works by utilizing an AI model and Nvidia’s Tensor cores on RTX graphics cards. Nvidia trains an AI model by comparing low-resolution, aliased images rendered by the game engine with a high-resolution reference image. After training, the model is bundled into a GPU driver, which can run the AI model in real time during gameplay using the Tensor cores’ computational power.

Nvidia DLAA diagram

To better understand DLAA, it’s important to revisit TAA. TAA collects one sample per pixel and jitters the pixels to gather more information for an average color value. However, TAA struggles with motion, leading to ghosting effects. DLAA solves this problem by leveraging the AI model to track motion, lighting changes, and edges in the scene, resulting in a cleaner image without ghosting.

DLSS and DLAA work in a similar way, with the main difference being the emphasis on performance and image quality. DLSS focuses on producing acceptable image quality with a significant performance gain, while DLAA prioritizes providing the best image quality at a performance loss.

Nvidia DLAA in Action

DLAA vs. TAA in The Elder Scrolls Online

DLAA can be seen in action in various games. In The Elder Scrolls Online, DLAA provides similar image quality to TAA, but without the ghosting and smearing issues that TAA occasionally exhibits. The difference becomes more pronounced at lower resolutions, where DLAA’s AI model can work wonders.

In Marvel’s Spider-Man, DLAA manages to extract more detail from distant objects compared to TAA, making it a suitable option for games with vast landscapes. On the other hand, in a game like Baldur’s Gate 3, DLAA doesn’t offer significant improvements over TAA in static images, with the quality depending on the game’s speed.

List of Games Supporting Nvidia DLAA

DLAA has been added to a growing list of games. Some of the notable titles that feature DLAA include:

  • A Plague Tale: Requiem
  • Baldur’s Gate 3
  • Call of Duty Modern Warfare II
  • Call of Duty Warzone 2.0
  • Cyberpunk 2077
  • Diablo IV
  • Marvel’s Spider-Man

and many more.

The Distinction between DLSS and DLAA


It’s essential to differentiate between DLSS and DLAA. Although they work with the same underlying technology, they serve opposite purposes. DLSS focuses on performance optimization at the cost of image quality, while DLAA prioritizes image quality at the cost of performance.

DLAA finds its usefulness in games like The Elder Scrolls Online, where players with extra GPU power can benefit from enhanced visuals. However, DLAA is limited to Nvidia’s RTX 2000 and RTX 3000 graphics cards due to its reliance on Tensor cores, which might exclude users with older hardware.

In conclusion, Nvidia DLAA brings a new level of image quality to gaming by leveraging AI models and advanced anti-aliasing techniques. Although it may come at a performance cost, the visual enhancements are worth it for players seeking the best possible experience. With an expanding list of supported games, DLAA is set to become a valuable tool for gamers who prioritize image quality in their gameplay.