A machine learning feature – DLSS feature could totally change your game experience.
The most-spoken-about abilities of Nvidia’s GeForce RTX graphics cards are Ray Tracing and 4K. There’s not a lot of hype about the DLSS feature these GPUs can use, but it’s also amazing, particularly because of the magic that it can operate on frame rates for those hardware players which in mid-range and low range, that can make you play games much faster. Here is a brief look at DLSS, how it functions, and the games support it.
DLSS (Deep Learning Super Sampling) is a machine learning feature used in Nvidia’s RTX 20 and 30 series graphics cards using AI Tensor cores. Your GPU renders in-game graphics at a lower resolution when DLSS is turned on, then uses AI to artificially upscale and upgrade the image to a higher resolution without losing performance. Nvidia has “trained” its DLSS algorithm to accomplish this with a supercomputer specially designed for graphics rendering.
For instance, DLSS renders the game at 1440p instead of rendering the game at 4K, after that the machine learning algorithms should work. The result is that the picture looks almost identical to the native 4K resolution, but runs as if it were viewed at 1440p, which can improve your framerate in-game and leave plenty of ray tracing and other high-end graphics settings device resources.
DLSS definitely change game outlook; it enables even the weakest RTX cards from Nvidia, such as the RTX 2060, to run at a compelling (and playable) 4K resolution, even though hardware is usually unable to work that way. Too bad, DLSS is not supported by PlayStation 5 and Xbox Series X; yes, they may be capable of 4K and raytracing, but turning these settings on always results in your framerate, forcing the player to choose between graphic fidelity or consistent output. PC players with a device that is compatible with DLSS do not have to pick.
The DLSS, however, is not perfect; this role is known to cause textures to be blurred and to lose some detail. While DLSS 2.0, the latest iteration, fixes many of these problems and increases the 4X resolution (1080p to 4K), you may experience some artifacts and blurring, particularly in the finer details. But since the DLSS 2.0 algorithm is constantly being revised, only over time can this function get better.
Another drawback of DLSS is its usability; it is only available, at least for now, in a small number of games. Nvidia had to train DLSS on a per-game basis initially, but the DLSS 2.0 algorithm can apply its improvements to any game that supports this function and makes it simple to implement for developers.