Nvidia introduces AI-enhanced DLSS 3.5 for all RTX graphics cards

DLSS 3.5 has also been trained on five times more data than DLSS 3, and will be available as a separate option in supported titles. (Image: Nvidia)

American chip designer Nvidia has introduced the next big update for its frame generation technology, DLSS 3.5.

DLSS, or Deep Learning Super Sampling, is an artificial intelligence (AI)-based image upscaling technique that allows games to run at a lower resolution for performance but upscales the image quality at roughly the same detail as if it were running at a higher resolution.

DLSS 3.5 promises to boost frame rates in ray tracing applications. Nvidia is targeting an advanced technique known as path tracing or full-ray tracing.

In graphics, ray tracing is a technique used to model how light would interact in a 3D world. Path tracing takes it a step further to improve global illumination based on real-world lighting principles.

Also read | Nvidia reports record Q2 earnings powered by AI boom: Key highlights

Since these are both expensive techniques, they are typically used economically within a game's pipeline and require powerful hardware at the consumer-end to experience it fully.

Also read | Nvidia’s stock surge pushes CEO’s fortune as high as $46 billion

With DLSS 3.5's AI-based Ray Reconstruction technique, Nvidia is promising a major boost in games that use the path-tracing pipeline. With the new update, Nvidia will now use an AI-powered network to hide small imperfections like minor losses of color accuracy, or ghosting.

DLSS 3.5 has also been trained on five times more data than DLSS 3, and will be available as a separate option in supported titles.