{"id":82640,"date":"2021-06-22T18:58:19","date_gmt":"2021-06-22T13:58:19","guid":{"rendered":"https:\/\/www.thenerdmag.com\/?p=82640"},"modified":"2021-06-28T13:55:24","modified_gmt":"2021-06-28T08:55:24","slug":"amd-fidelityfx-released-a-worthy-rival-for-nvidia-dlss","status":"publish","type":"post","link":"https:\/\/thenerdmag.com\/amd-fidelityfx-released-a-worthy-rival-for-nvidia-dlss\/","title":{"rendered":"AMD FidelityFX released – a worthy rival for Nvidia DLSS?"},"content":{"rendered":"\n

About a year ago, AMD <\/strong>had announced its FidelityFX Super Resolution (FSR)<\/a><\/strong> technology. This technology aimed to provide a boost in performance for users while gaming and was a reply to Deep Learning Super Sampling (DLSS)<\/a><\/strong> by Nvidia<\/strong>. While Nvidia<\/strong>‘s DLSS <\/strong>used Deep Learning and AI to function, AMD <\/strong>said that their FSR <\/strong>technology doesn’t do that stuff.<\/p>\n\n\n\n

Perhaps what no one saw coming was the fact that AMD <\/strong>not only enabled FSR <\/strong>technology on their own AMD <\/strong>Graphic Cards but also Nvidia <\/strong>Graphic Cards, all the way to the Pascal <\/strong>series and showed various benchmarks in various games in terms of FSR <\/strong>performance (Read more here<\/a>). They aimed to advance their own technology and at the same time shared it with others too. What a chad move! But they did they bring out something good to the table? <\/p>\n\n\n\n

How does FSR work?<\/h2>\n\n\n\n

So how does FSR <\/strong>actually work? Well, according to a patent, “The combination of the linear and non-linear upscaling facilitates both the preservation of color and larger-scale features (large objects and shapes that are more easily perceived by the human eye) of the image from linear upscaling as well as the preservation of finer features (e.g., curved features and features that are not easily perceived in low resolution) of the image from non-linear upscaling.”<\/em><\/p>\n\n\n\n

What does all this mean? There are two networks, a linear <\/strong>upscaling network and a non-linear<\/strong> upscaling network through which the low resolution frame goes at the same time. The results are combined into a grid of pixels and generates a frame with higher resolution pixels. This requires no training or machine learning.<\/p>\n\n\n\n

\"AMD