DLSS (Deep Learning Super Sampling)
TLDR: DLSS (Deep Learning Super Sampling) is an AI-driven rendering technology developed by NVIDIA and introduced in 2018 alongside the GeForce RTX 20 series GPUs. It uses deep learning to upscale lower-resolution images in real-time, producing visuals that rival native resolution while improving performance. By offloading rendering workloads to dedicated Tensor Cores, DLSS enables smoother gameplay and higher frame rates in graphically demanding titles.
https://en.wikipedia.org/wiki/Deep_learning_super_sampling
DLSS leverages neural networks trained on high-resolution image data to reconstruct missing details and sharpen frames rendered at lower resolutions. This process reduces the computational burden on the GPU, allowing games to run at higher settings without sacrificing performance. DLSS has evolved through multiple versions, with DLSS 2.0 offering improved visual fidelity and broader compatibility, and DLSS 3.0 introducing frame generation for even greater performance gains.
https://www.nvidia.com/en-us/geforce/technologies/dlss/
The integration of DLSS in modern games enhances the gaming experience by balancing visual quality and performance, especially for titles utilizing ray tracing and other resource-intensive effects. Supported by NVIDIA RTX GPUs, DLSS is a key feature in achieving high-refresh-rate 4K gaming and making VR experiences more immersive. Its ongoing development ensures compatibility with new games and hardware, solidifying DLSS as a cornerstone of NVIDIA's GPU ecosystem.