Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now


Beamr, Nvidia partner on 10x faster encoding solution

Integrated solution enables video providers to optimise video files and live video streams in real time

Video encoding company Beamr is launching an integrated solution that combines its CABR (Content Adaptive BitRate) rate control with the NVIDIA NVENC hardware video encoder.

According to Beamr, the new solution is 10x faster for the same encoding quality.

The integrated solution enables video providers to optimise video files and live video streams in real time, improving the user experience and reducing storage and delivery costs, said the company.

Beamr added that its CABR engine can reduce the bitrate of video streams by up to 50 per cent. The reduced-bitrate streams remain fully standards-compliant, and can be decoded by any standard video decoder.

NVIDIA has added a frame-level application programming interface, or API, which enables Beamr’s CABR to determine the optimal tradeoff between bitrate and quality and “significantly reduce” the bitrate of video streams.

Beamr has also refactored its CABR technology to run on NVIDIA GPUs at ultra-fast performance, matching the performance of NVIDIA NVENC, enabling 10x more cost efficiency than software-based content-adaptive encoding, said the company.

“Beamr has been optimising images and videos for the past 10 years, resulting in reduced storage and delivery costs, and an improved user experience for our customers,” said Sharon Carmel, founder and  CEO of Beamr.

“The increase in video quality with the shift to 4K and 8K, and the increase in the amount of video with applications such as smart cities, automotive and user-generated content, require new levels of video optimisation performance, which are available only with hardware encoders. We selected the NVIDIA NVENC hardware video encoder due to its high quality and extreme performance, and are grateful to the NVIDIA team for working closely with us on adapting the NVENC APIs to enable frame-level access to the video encoding engine.”