RTX Video Super Resolution Version 1.5 brings improved image quality and de-artifacts content

Earlier this year, NVIDIA launched RTX Video Super Resolution for GeForce RTX 30 and 40 Series GPUs, which is like a version of DLSS for video streaming on platforms like YouTube - take a lower-resolution video and use AI to upscale it to your native resolution. Like DLSS, it uses the Tensor Core AI hardware in GeForce RTX graphics cards.

NVIDIA's RTX Video Super Resolution updated to Version 1.5.

That was back in February, and now, after several more months of training the RTX VSR AI model, NVIDIA is launching RTX Video Super Resolution or RTX VSR version 1.5 - now available as part of the latest GeForce Game Ready Driver. And support is expanding to the entire GeForce RTX family - including the Turing-based GeForce RTX 20 Series.

So, what's new? Alongside improvements to image quality, there's a new option that will de-artifact video played at your display's native resolution.

Based on the screenshot supplied by NVIDIA, which showcases a 1080p stream played in its original quality next to an RTX Video Super Resolution enhanced version - there are clearly fewer artifacts. It's impressive, and it looks like NVIDIA is serious about bringing AI-based video upscaling and image processing to the PC space for services like YouTube, Twitch, Netflix, Disney+, and other streaming platforms.

At launch, RTX VSR was an impressive bit of technology; however, like the original version of DLSS, there was room for improvement. "Retraining the VSR AI model helped it learn to accurately identify the difference between subtle details and compression artifacts," writes NVIDIA. "As a result, AI-enhanced images more accurately preserve details during the upscale process. Finer details are more visible, and the overall image looks sharper and crisper."