It would seem as if Nintendo has been investigating a wide variety of alternative options in order to enhance the visual quality of the Nintendo Switch as the device continues to age. Since its release in March 2017, when the Nintendo Switch first went on sale, the console’s internal hardware has, for the most part, stayed unaltered.
Improved image processing
The business submitted a patent application in June 2022, and it seems to go some distance toward helping to enhance the picture quality of the device. Since one of the inventors for the patent is Akashi Miyagi, who has previously worked as a sub-director with Kentaro Tominagaon on The Legend of Zelda: Twilight Princess and is thought to play a role in the upcoming The Legend of Zelda: Tears of the Kingdom, it is possible that the patent is related to the upcoming The Legend of Zelda: Tears of the Kingdom. This is due to the fact that Akashi Miyagi is listed as one
“An image processing technique for expressing objects with transparency while maintaining a low processing burden,” reads the description of the patent.
And this does not seem to be Nvidia’s DLSS
NVIDIA DLSS is credited for revolutionizing the graphics industry by using AI super resolution and Tensor Cores on GeForce RTX GPUs in order to increase frame rates while producing visuals of superior clarity and quality that are competitive with those produced by native resolution. Since Deep Learning Super Sampling (DLSS) was made available, 216 different games and applications have integrated the technique. This has resulted in quicker frame rates and the performance headroom necessary to make real-time gaming ray tracing a possibility.
NVIDIA DLSS 3, the next revolution in neural graphics, has been officially announced today, and we couldn’t be more thrilled. DLSS 3 increases speed by up to four times more than brute-force rendering by combining DLSS Super Resolution, the all-new DLSS Frame Generation, and NVIDIA Reflex. This is made possible by the enhanced hardware capabilities of GeForce RTX 40 Series GPUs. Over 35 games and apps have already begun incorporating the technology, with the first of them launching in October. The ecosystem is moving quickly to implement DLSS 3, which is already seeing fast adoption.
The Development of NVIDIA DLSS in Recent Years
When we initially presented NVIDIA DLSS, our goal was to revolutionize real-time rendering using AI-based super resolution. This included producing fewer pixels and then use AI to generate pictures with a sharper and higher resolution. A little over one and a half years later, we presented the world with NVIDIA DLSS 2, which significantly enhanced picture quality and performance with a generic neural network that could adapt to any game or setting without the need for any kind of specialized training. DLSS 2 is currently supported in both Unity and Unreal Engine, and it has been integrated into 216 different games and applications thus far. DLSS 2 keeps getting better because to continual training on NVIDIA’s artificial intelligence supercomputer. To far, there have been 4 significant upgrades issued, each of which has brought about further improvements to the picture quality.
A Revolution in Neural Graphics: Introducing NVIDIA DLSS 3
DLSS 3 is a groundbreaking innovation in AI-powered graphics that substantially improves speed while preserving excellent picture quality and responsiveness. It was developed by Nvidia and Microsoft. Building upon the foundation laid by DLSS Super Resolution, DLSS 3 includes the addition of Optical Multi Frame Generation, which generates totally new frames, as well as the integration of NVIDIA Reflex low latency technology, which ensures the best possible responsiveness. GeForce RTX 40 Series graphics cards are powered by NVIDIA’s latest iteration of Ada Lovelace architecture, which includes the fourth-generation Tensor Cores and Optical Flow Accelerator. DLSS 3 is driven by these components.
The current and previous game frames, an optical flow field created by Ada’s Optical Flow Accelerator, and game engine data such as motion vectors and depth are the four inputs that the DLSS Frame Generation convolutional autoencoder receives.
