The State of Video Streaming in 2013
Let’s say you publish DVDs for sale.
Disc replicator A says they can provide DVDs that work for 99.4% of customers.
Disc replicator B can provide DVDs that work for 65% of customers.
Imagine the hassle of customer complaints and refunds if you went with disc replicator B!
If the costs were close, the choice of which replicator to go with would be very clear. In fact, even if disc replicator A charged 50% more, you’d come out way ahead not having to deal with the extra customer service costs.
But while this logic is clear when it comes to DVDs, there’s a lot of confusion when it comes to internet-delivered video.
In a nutshell, if you don’t want to read further, I’ll just say that Leaping Brain Labs is replicator A. Streaming-only providers are replicator Bs.
Now, for the details:
Akamai just published their “State of the Internet” report yesterday with valuable information for those of us in the field of internet delivery of video. Herein I will address the current state of delivery of streaming video.
Theoretically, under ideal conditions you could view a 1.5 Mbps HD video stream with a solid 2 Mbps connection. But only if your connection is pristine, not shared, you are using Ethernet, your ISP is exceptional and you have a late model laptop or other device with fast networking.
In reality, however, you will rarely find ideal conditions like these.
The truth is, to deliver a quality fullscreen HD experience, you need a consistent connection of at least 3x to 4x the bitrate of the video you’re delivering.
Why 3x to 4x?
The additional bandwidth required is called “overhead”. Internet video is delivered in packets that are constantly hopping around on the tubes of the interwebs, meaning extra last mile bandwidth is needed to sustain playback when internet congestion delays delivery of data packets.
If someone else in the household is playing Halo with friends, watching YouTube, downloading MP3s or even surfing the internet on their iPad, the total bandwidth to that household is divided between all the connections requesting it, usually by a several-year-old $79 wireless router, not exactly enterprise-grade equipment. Wireless networking itself has its own overhead requirements and bottlenecks. And individual devices, from Google TVs to laptops to smartphones each have processing and bandwidth overhead.
Most HD video, depending on the content, can be encoded in fairly decent quality at anywhere from 1.2 – 4 Mbps – if you are expert and using world class tools like SampleLab in DV Kitchen.
So, as a benchmark, by the authority vested in me as a respected internet video pioneer and expert, I will say you need a solid, sustained connection of a bare minimum of 4 Mbps to have an acceptable experience viewing decent quality streaming HD video over the internet.
So how many people have a connection this fast?
The Akamai report shows that in the U.S. (and Great Britain), only 65% of connections surpass this mark.
In other fertile media markets, such as Ireland, France and Spain, about half pass the test. Only 1 in 3 of our friends down under qualify. And in important emerging markets like Thailand (25%), Brazil (13%), China (5%) and India (3%), it’s clear that streaming video is not an option.
Also remember that ISPs go down and whole areas lose internet connections. People’s wireless routers sometimes break. Servers go down on a regular basis, even Amazon’s. All this flakiness means one thing:
It is not a smart plan to rely on video streaming as the only way your customers can access your video content.
The Leaping Brain Fido™ downloading library ensures media delivery even to emerging markets with highly intermittent connections, and perfect fullscreen playback even if the user’s net connection goes out entirely.
Bottom line: If you want to 99.4% of your customers to have an excellent video consumption experience, the Leaping Brain platform is your only choice.
Check out the map here.