Nvidia DLSS 10 predicted to deliver a complete neural rendering system

We're now at a point where AI technology and DLSS can produce image quality that is noticeably better than native rendering, thanks to the arrival of DLSS 3.5 and the…

By Nivedita Bangari

September 26, 2023

2 min read

Comments

We’re now at a point where AI technology and DLSS can produce image quality that is noticeably better than native rendering, thanks to the arrival of DLSS 3.5 and the new AI-powered Ray Reconstruction technology, which produced some genuinely game-changing results in Cyberpunk 2077’s Update 2.0 and Phantom Liberty expansion.

As bold as that statement may be, it refers to Cyberpunk 2077’s AI-powered Ray Reconstruction, which has the full RT Overdrive Mode enabled.

With DLSS 3.5, an image is significantly less noisy, the quality of reflections is enhanced, ambient occlusion detail is clearer, and RT lighting effects are improved.

The holiday season sales that e-commerce businesses run from September to October are essential because they help gauge consumer sentiment, which in turn affects how well businesses operate. In total, sales during the holiday season comprise around 50% of the gross merchandise value (GMV) for e-commerce businesses.

As long as consumer confidence is positive, holiday sales are expected to increase by up to 20% year over year (YoY), reaching Rs 90,000 crore in GMV this year, according to Redseer analysts.

Bryan Catanzaro, Vice President of Applied Deep Learning Research at NVIDIA, is participating in a fascinating roundtable discussion hosted by Digital Foundry and is already considering the future of DLSS, including how a hypothetical DLSS 10 will be a full “neural rendering system that interfaces with a game engine.”

A whole system would imply that AI is handling and presenting everything you see on screen, similar to Frame Generation, Ray Reconstruction, and other rendering capabilities and technologies that are currently a part of NVIDIA DLSS.

Bryan Catanzaro brought up a driving demo the business created in 2018 during the meeting. A neural network that handled “every part of the rendering process” was fed information from a gaming engine, such as where objects were situated and what was happening. Thanks to AI, where a whole scene may be built from input alone, 3D environments may become more realistic and detailed.

Also Read:

Source