Nvidia’s optical streaming accelerators can use AI frame generation for video encoding, too

Something to look forward to: Many consumers are looking forward to this year’s new graphics cards for their gaming capabilities, but they are also introducing new video encoding tools. Nvidia’s RTX 4000 GPUs add another trick to doubling frames while encoding video.

The optical streaming accelerators behind Nvidia’s new DLSS 3 feature will not only increase video game frame rates. Content creators can also use technology artificially Increase framerates In the videos they encode.

While DLSS 2.0 uses Tensor Cores in Nvidia’s RTX 2000 and 3000 GPUs to create new pixels through machine learning, DLSS 3 Uses 4000 Series Optical Flow Accelerators for all-new tire construction. PC games that support DLSS 3 can double their frame rates as well as the performance gains of DLSS 2.0, but Nvidia’s technology can bring the same improvements to videos.

Motion vectors are one of the tools that DLSS uses to improve game frames, and Nvidia also uses it in what it calls an engine-assisted frame rate conversion (FRUC). Essentially, it is a form of hardware-assisted kinematic interpolation. The concept is similar to how TVs smooth and interpolate motion, but the RTX 4000’s CUDA cores and optical flow accelerators make the process faster and more accurate. When interpolated frames contain artifacts, image field hole-filling techniques can fill them in to create an accurate final image.

The FRUC library APIs support ARGB and NV12 input surface formats. It is also compatible with all DirectX and CUDA applications.

Improved kinetic interpolation can distinguish Lovelace from Intel’s Arc Alchemist series and AMD RDNA3 GPUs as all three offer AV1 GPU-based encoding. Early tests showed that AV1 has significant advantages over H.264 in terms of speed, data usage, and image quality. The new format enables content creators and content creators to encode HD videos more efficiently. Unlike H.265, AV1 is also royalty-free.

Google is also pushing the AV1 codec as the format is becoming increasingly important on YouTube. This week, the company chest A significant update to the open source AV1 encoder – AOM-AV1 3.5 – which now supports frame-parallel encoding for more threads. Depending on the video resolution and the number of processor threads, the update can reduce encoding times by between 18 and 34 percent.

Source link

Related Posts