Deep Learning Super Sampling (DLSS)
Deep Learning Super Sampling (DLSS) is a revolutionary AI-driven upscaling technology developed by NVIDIA, first introduced in 2018 alongside the GeForce RTX 20 Series GPUs. DLSS leverages the power of Tensor Cores present in NVIDIA’s Turing architecture and later architectures like Ampere and Ada Lovelace. By using deep learning models, DLSS renders games at a lower resolution and then upscales the image to higher resolutions, providing improved performance without compromising visual quality. This process reduces the computational burden on the GPU while delivering crisp and detailed graphics. https://en.wikipedia.org/wiki/Deep_Learning_Super_Sampling
DLSS enhances gaming experiences by enabling real-time ray tracing and high-resolution gameplay without a significant loss in frame rates. Versions like DLSS 2.0, introduced in 2020, refined the technology by improving image quality and supporting a wider range of games. This version uses a generalized AI model that works across multiple titles, reducing the need for game-specific training. The technology has been successfully implemented in titles like Cyberpunk 2077, Control, and Battlefield 2042, showcasing its ability to deliver stunning visuals while maintaining smooth performance. https://www.nvidia.com/en-us/geforce/technologies/dlss/
In 2022, NVIDIA introduced DLSS 3 with the launch of the GeForce RTX 40 Series, adding support for AI-generated frames. This innovation further enhances performance by interpolating additional frames, making it particularly effective for graphically demanding scenarios. DLSS 3 not only improves frame rates but also reduces latency, benefiting competitive gaming. As a core feature of modern NVIDIA GPUs, DLSS continues to push the boundaries of visual fidelity and performance in gaming and professional applications. https://www.nvidia.com/en-us/geforce/rtx-40-series/dlss-3/