NVIDIA DLSS explained: How AI upscaling boosts game performance
NVIDIA's DLSS is a proprietary upscaling technology that lets a GPU render a game at a lower internal resolution and then uses a machine-learning model to scale the image to the display's native resolution. When implemented well, it can improve performance enough to allow higher graphical settings or higher frame rates. The trade-off is some loss of image fidelity, with the amount depending on the selected DLSS mode such as Quality.
DLSS is NVIDIA’s exclusive image-reconstruction system. A compatible GPU draws fewer pixels first, then a trained neural model enlarges the frame to the chosen output size, ideally the panel’s native resolution. The model uses motion, depth, and luminance cues to estimate missing detail.
Presets set the starting resolution. Quality begins at 66.7 percent of output, Balanced at 58 percent, Performance at 50 percent, and Ultra Performance at 33.3 percent. Quality preserves the most detail; lower tiers trade clarity for speed. Ultra Performance targets 8K, while 1440p or 4K users may see blur.
DLSS could affect how gamers, developers, and hardware makers approach performance. Players may gain smoother frame rates without immediate GPU upgrades, while developers could rely more on upscaling to hit visual targets. This may make image-quality trade-offs more routine and could influence purchasing decisions, though results will vary by game, display, and preset.