Fast AV1 Encoding Is Finally Here
Nathan Veer
Director of Product Marketing
For years “codec complexity” ergo “high computational cost” has hung over the AV1 codec, causing some to hold off adopting the standard. Since video codec developers are continuously improving the performance of their codec implementations, this assumption doesn’t have to be true.
The secret to breaking the operational cost barrier for AV1 is to use an encoder that was built from the ground up, to take full advantage of all of the incredible encoding tools AV1 has to offer. Through joint core RDO optimization and adaptive AI-based video processing, we pull all those levers and squeeze every ounce of quality out of every bit for you.
At Visionular, we ran tests using our Aurora1 AV1 encoder on the new AMD EPYC 7763 processor and were able to outperform SVT-AV1 by a factor of up to 7X.
That’s right, we built a race car of an encoder that uses less fuel (if you will) in a world of “bicycle encoders”. Check it out for yourself with these side-by-side comparisons.
Continue reading...
At Foothill Ventures, we believe in startup companies that ride the transformative power of major technology shifts such as deep learning in computer vision. Visionular’s founders are world-class technologists in their field of video codec and AI-driven optimization. We feel privileged to support their adventure with our resources and experience.
I invested in Visionular because the team is at the forefront of innovations in video encoding and image processing for real-time low latency video communications and premium video streaming applications.
