It seems that Sony is thinking about implementing a similar solution in its Playstation 5, as it has patented a similar rescaling technology using artificial intelligence. With the launch of the GeForce RTX graphics cards, Nvidia announced its DLSS technology, an acronym for Deep Learning Super-Sampling, which allows images to be rendered at a lower resolution than desired and then rescaled to the final resolution by artificial intelligence, resulting in higher FPS thanks to the rendering of lower resolution images.
This is what the patent in question says:
“Information processing apparatus for obtaining a plurality of reference images by plotting an object to be reproduced, obtaining a plurality of converted images by enlarging or reducing each of the plurality of reference images, performing machine learning using a plurality of images to be learned as learning information including the plurality of converted images, and generating pre-learned information used to produce a reproduction image representing the appearance of the object.”
While Sony’s spelling is somewhat confusing, it basically means that you have a device that renders images and scales them to both larger and smaller resolutions and then uses them to train an AI. Then that AI creates images based on what you learned, which would basically be frames from the video game.
This is very similar to what Nvidia does and could allow the PS5 to reach 60FPS in 4K thanks to the use of AI and not just raw playback. So far, Sony hasn’t officially mentioned anything when it came to the Playstation 5, so it’s not clear whether it will be used in the console or not, although we’ll certainly have news soon if it is.