Vidyo.io SFU Delivering VideoPeople listening to Vidyo will often hear us discuss differentiators such as high reliability, dynamic optimization, error resiliency, and so on. These are all key benefits of the architecture that Vidyo has been diligently developing since our founding in 2005. This core architecture has allowed us to deliver a truly enterprise-grade and diverse video conferencing offering that has delighted our enterprise customers for years.

Video conferencing technology was once only reserved for experts. Developing a system for users to be able to have real-time, interactive, multiparty video communication required specialized knowledge. We have been developing and refining our innovative architecture over the last dozen years into an enterprise-grade video conferencing system that works reliably over low-quality networks. We built Vidyo.io to put this technology in the hands of any developer with a great idea who wants to  take advantage of our platform, and embed video chat into their own apps. We made Vidyo.io very easy for developers to work with; in fact, a developer does not need any specialized video knowledge to use Vidyo.io. They can simply use the APIs and embed video in a matter of minutes.

So how does it work? Vidyo.io leverages Scalable Video Coding (SVC) technology, combined with our patented Selective Forward Unit (SFU) infrastructure, to dynamically optimize the video that is transmitted to each participant in the call. The SVC encoded video stream allows the stream to vary in both frame rate and resolution. The SFU continuously assesses each endpoint’s need based upon the endpoints’ CPU, screen resolution, and available bandwidth. These parameters are used by the SFU to determine how much, or how little, of the resolution and frame rate information to forward to each of the endpoints. The result is a uniquely optimized video experience for each endpoint in the call. Each endpoint receives the highest quality video chat experience possible. The benefit of this architecture is an extremely reliable call, even over poor quality networks, such as wireless.

How is your Selective Forwarding Unit different from others? Vidyo certainly was the first to develop an SFU and make it commercially available. One major difference between the Vidyo.io SFU and other SFUs is our use of SVC technology in conjunction with an SFU.

Many other platforms use a technique known as simulcast in order to provide resolution diversity for connected endpoints. For example, if there are two connected endpoints in a call each requiring a different resolution, the sending endpoint has to encode and transmit two different video streams. In contrast to SVC, simulcast puts the burden on the sending endpoint – which can be particularly challenging for video collaboration over mobile devices.

Simulcast SFU

The Vidyo.io SFU uses Scalable Video Coding techniques for encoding the video with both temporal and spatial scalability. The SFU, in this case, is aware of the spatial (resolution) and temporal (frame rate) layers and can thin the video stream based upon endpoint need. So if an endpoint connected can handle the full resolution and frame rate of the source video stream, the SFU forwards the full stream. However, if an endpoint needs a subset of the resolution and/or frame rate, the SFU can forward a subset of the layer information. Because the video is encoded with both temporal and spatial scalability, the SFU doesn’t need to ask the sending endpoint to transmit a new stream. Also, the Vidyo.io SFU can very precisely size the outgoing video stream based upon endpoint need. This ensures each endpoint receives the optimal video quality possible, and that each video call participant has a fantastic user experience.

Vidyo.io SFU

What codecs support SVC? The beautiful part about SVC is that it is a technique for encoding video. It is not tied to a specific codec. As a result, Vidyo has been actively involved in the development of various codecs – we’ve helped provide SVC to H.264, H.265, VP9, and we are working with the Alliance for Open Media on their upcoming royalty-free codec. This means that the Vidyo.io SFU can easily be adapted to support a variety of codecs while still delivering the same adaptive video optimizations.

Try it for yourself! The Vidyo.io site has a demo app that is ready to go. You can try the adaptive video technology without needing to write any code. All you need to do is sign-up for an account and check out the Vidyo.io demo. So go SEE it for yourself.