AMD’s new Redon RX 90 Series cards and RDNA 4 architecture formally launch on March 5, and a new version of the AMD’s Feltifil FX Super Resolution (FSR) is coming along.
FSR and NVDIA’s deep learning super samples (DLSS) have the same goal of the Apostlers: Taking a low resolution image presented by your graphics card, colliding the resolution, and making a locally provided pixels to create an image. Turn off GPU to perform work without any work. They can make advanced mistakes, and they will not always look good as a local resolution image. But they are both good alternatives to live with fading, non-local-resolution pictures on the LCD or ON LED display.
FSR and DLSS are especially useful for old or cheap 1080p or 1440p-capable GPUs, which are connected to the 4K monitor, where you have to decide between otherwise sharp 4K image and sports frame rate. It is also useful for hitting high frame rates on low resolutions, which can be useful for a high refresh rate gaming monitor.
But unlike the past version of the FSR, the FSR4 is promoting photos using hardware -backed machine learning algorithm, the hardware has been newly added to the RDNA 4 and RX90 series graphics card. It is a mirror of NVIDIA strategy with DLSS, which has always taken advantage of tensor cores found in RTX GPUS to run machine learning model to achieve high image quality for high and AI-generated frames. If you do not have RDNA4 GPU, you cannot use FSR 4.