NGCodec, a company focused on FPGA-based H.265/HEVC video codec development, today announced that it has been working with Amazon Web Services (AWS) in the development of RealityCodec™ for hardware-accelerated AWS-based video encoding and VR / AR processing.
RealityCodec allows for AWS-based, ultra-high-performance video compression (at up to 4K resolution) and ultra-low sub-frame latency cloud-based VR and AR (should this feature be implemented by Amazon) for all AWS customers. By running the compute function in the cloud, it is possible to send only the graphics pixels in the form of a video feed to the HMD (Head Mounted Display). The benefits to this approach are higher performance, lower cost and longer battery life for the consumer VR/AR hardware.
NGCodec’s RealityCodec technology runs on NGCodec-designed FPGA instances, presenting significant advantages for AWS cloud computing compared to traditional software encoding, including:
“AWS is breaking ground as the first public cloud to add FPGA instances, which offers significant acceleration over CPU and GPU instance types,” said Oliver Gunasekara, CEO & co-founder, NGCodec Inc. “Many large media workflows, including on-demand and live streaming content, are moving to the AWS Cloud. These can now have all the benefits of FPGA acceleration in the AWS ecosystem.”
Optimized for the newly-announced F1 AWS FPGA instance type, NGCodec will present a demonstration using Cloud FPGA acceleration for live H.265/HEVC video encoding on December 1, 2016 at the AWS re:Invent 2016 event in Las Vegas, Nevada.
“We have hand-selected a small number of companies to demonstrate FPGA acceleration at AWS re:Invent 2016,” said [name], [title] at Amazon Web Services. “In just three weeks, NGCodec, a leader in video compression, was able to port its FPGA encoder to run on the AWS Cloud using the new F1 instance type. We are looking forward to working together to enable these FPGA benefits to AWS Cloud customers.”