ASUS Gaming GeForce RTX 4070 TUF review

Graphics cards 1049 Page 19 of 31 Published by


DLSS3 performance


In conditions where you'd like (or need) a bit more FPS you can apply DLSS3 and Frame generation. DLSS, or Deep Learning Super Sampling, is a technology developed by NVIDIA that uses deep learning and artificial intelligence to upscale lower resolution images to higher resolution ones. DLSS 3.0 is the third iteration of this technology, featuring several improvements over its predecessors. DLSS 3.0 improves the image quality, performance, and efficiency of the upscaling process. It uses AI-based neural networks trained on a vast database of high-quality, high-resolution images from various games. The neural networks are then used to generate and predict high-resolution images by intelligently analyzing the lower resolution input.

DLSS Frame Generation
To generate a new frame, DLSS 3 takes the current frame from a game or application, analyzes it to determine the best way to upscale it, and then generates a new, higher resolution version of the frame using the trained AI model. This new frame is then displayed on the screen, resulting in a higher quality, smoother, and more detailed image. One of the advantages of DLSS 3 frame generation is that it can improve the performance of games on lower-end hardware by allowing them to run at lower resolutions while still producing high-quality images. This is because DLSS 3 can upscale the lower resolution image to a higher one that looks almost as good as a native resolution image, while using fewer resources to render the lower-resolution image.

Please note that the result set below is based on the reference review and serves as an indication of the performance of cards in the same class.




Share this content
Twitter Facebook Reddit WhatsApp Email Print