Bestfrontseat Comes Out of Stealth Mode With Advanced Compression

Since 2006 Bestfrontseat has been busily refining a turnkey live event video distribution service using a unique compression technology derived from medical imaging. As the UK-based company emerges from stealth mode, founder and CEO Steve John explains why he thinks he has a world-beating technology.

“We set the company up in June 2005, did some market analysis by inventing a fictitious live event, and attempted to engage companies to broadcast it for us,” explains John. “We discovered that there was no one system able to attend a live event and broadcast that event in real time. They could provide us with a codec or a server, a storage or web construction, but you’d have to engage multiple companies, at substantial cost each time, and there was no accountability if the broadcast went belly up.”

Content delivery networks "still are very expense to use, especially for one-off events or for annual events,” he says. “The high cost of data transfers almost always destroys any business model when you combine the costs of establishing and operating a server cluster able to handle any volume of viewers, the construction and management of a website and the associated database build.”

Furthermore, John says, the provision of video on demand, usually accessed after the live event, was found to be cost prohibitive for all but a very few events. So Bestfrontseat, with funding from John, businessman Peter Kemp, and private investors in Switzerland and the U.S., set about designing its own infrastructure from the ground up.

First it organized a group of server clusters capable of being available to multiple clients while still providing a service to large audience numbers.

“The design included provision for all database needs to ensure secure and controlled access to the live and VOD content,” says John. “We linked those server clusters into a global CDN (there are three providers but he's not saying which) and also provided a web interface to enable the viewer to access the content under the brand of the content owner/promoter/artist.”

This was seen as a critical aspect of the development. Bestfrontseat’s white-label solution provides clients with their own branded front end.

“The high costs associated with data transfer rates made life very difficult, as even the best business model can fall apart when delivery takes more than 30% of sale price,” he says. “Factor in royalty payments and production costs and online video becomes an expensive luxury. The lower the bandwidth, the lower the delivery costs, but the lower the image quality.”

Proprietary Compression Technology
Bestfrontseat believes it has an ace up its sleeve in the form of a compression system it acquired from a medical scanning research project—a place where John had previously worked.

“I left the project in 2003 but I always had in mind a means to apply the codec in other areas. When I came to examine the online video market a couple of years later I realized that there was no definitive solution and that the codec was a perfect fit.”

Using the PSI Video compression engine and progressive downloads rather than Real Time Streaming Protocol (RTSP), Bestfronseat claims to deliver near DVD-quality video content at a faction of the bit rates of all the other codecs.

“Today if you broadcast at near DVD-quality full screen you need 1.2Mbps (in H.264). When we do the same we need 550-650Kbps with 96Kbps for full stereo sound included within that bit rate, so in effect we are placing half the load on the local infrastructure.”

This, he claims, has dramatically reduced the delivery costs and also greatly improved the last mile issues.

“The lower the data stream rate the lower the end user’s connection can be to enable delivery without the buffering and download delays experienced with other compression delivery rates,” he says.

PSI Video works by treating what it sees more like the human eye than the traditional discreet cosine transform (DCT)-based approach and applies selective compression ratios to varying parts of the scene. This enables the codec to apply higher compression to less important aspects of a scene such as a static background element, and lower compression ratios to more visually important elements.