Home » Internet

Does Broadband really lag applications?

By George Ou 16 February 2010 8 Comments

Google is certainly getting a lot of media attention over their plans for an experimental gigabit broadband network.  The main argument for this type of a network is to give high bandwidth applications a home to be tested because the theory was that broadband networks in the US were constricting applications to very low bandwidth.  But does broadband really lag applications, or is it really the other way around?

Based on the latest Broadband data from the FCC, there were 86 million broadband customers by the end of 2008 in the United States.  Of those customers:

  • 56.67 million had more than 3 Mbps
  • 40.49 million had more than 6 Mbps
  • 13.59 million had more than 10 Mbps
  • 86 thousand had above 25 Mbps service

This data is more than a year old so we’re probably much faster now especially with free speed doubling upgrades from companies like Comcast due to competitive pressures from Verizon FiOS and AT&T U-verse.  But we’ll work with the 2008 data and compare it to application performance.

At the very end of 2008, YouTube was just barely bringing on 2.25 Mbps 720P video due to competitive pressures from Vimeo.  Even at the beginning of 2008, Google was still limping along at 0.32 Mbps content for YouTube and they soon upgraded to 0.64 Mbps content.  Broadband speeds had been more or less like this all of 2008 and not much slower on average in the previous two years.  The most popular content on the other hand was dramatically lower most of 2008 and a lot slower in the years prior.

By the end of 2009, YouTube had gone up to 3.75 Mbps 1080P content.  Other popular video sites like Hulu run at barely over 1 Mbps.  Only Pay content sites like Apple iTunes offer 4 Mbps content while Microsoft offers 6.9 Mbps content for their Xbox Live Marketplaces, but those aren’t used by the majority of users because the content is cost prohibitive.  Japanese content sites apparently use even lower bitrates due to the high cost of server hosting bandwidth in Japan, so all those fast broadband connections in Japan largely sit idle because the constraining factor is on the content side.

So YouTube is essentially ignoring the needs of more than 40 million American broadband users who have 6 or more Mbps of broadband performance.  Google only recently bumped up to 3.75 Mbps for their 1080P content which at times looks worse than standard definition DVD content for fast moving high complexity content because of the over compression.  If Google is so concerned about driving high performance broadband adoption, why don’t they offer 6, 8, 12, 20 Mbps content on a popular site like YouTube?

Many broadband customers can boost up from 1.5 to 6 Mbps broadband service for as little as $10/month extra, yet choose to remain on the 1.5 Mbps service for minor monthly savings.  But how can we blame them when the content has been lagging in performance?  I’ve known people who have upgraded their broadband plans once they have a favorite website that requires more bandwidth.  If Google wants to take a lead in driving higher speed broadband adoption, then why aren’t they bumping up the quality of YouTube?  What is the point of an experimental network with 20,000 homes when we already have over 40 million homes that aren’t getting the rich content that their connections are capable of?

8 Comments »

  • Twitted by DigiSociety said:

    [...] This post was Twitted by DigiSociety [...]

  • Michael Baumli said:

    One Word Answer: Never

    Currently I am writing this from what is considered the best internet connections in the world. Rarely does the application exceed the highest standards available in technology. This simply does not happen.

    Look at computer gaming for instance. We have what, one game in the last 20 years that actually exceeded the current specifications available in computing. We all remember Crysis, and the joke still exists, “Will it play Crysis?” The problem is that when a company provides a good that only a very select portion of the market can take advantage of, then the product can’t sell itself to pay for its own costs. Much like the cost of running fiber out to the middle of Nebraska so that John Smith can use broadband to teach John Junior, the cost is too high and only a single customer won’t cover the millions in cost to collect $70 or so a month they might get for the service.

    Is this a Chicken before the Egg scenario? I don’t really know. I hope that in time we will start to see streaming that better utilizes the broadband. Sorry, but even 1080P streaming leaves a lot to be desired because it compresses almost too much of the data. Then we can talk about high quality audio which I am sure could also push the limits a little bit. After years of pushing MP3 and AAC formats which teach our end users that crap is acceptable and true CD quality becomes a past thing, we might end up re-introducing high quality to the next generation after we learn we actually have space for high quality audio. This might take about another 5 years though because I know many friends of mine still talk about the days of using CDs and collecting MP3s at higher bit rates.

    In some ways there seems no reason to begin to start services that can utilize a full 100 mbps connection or better. As you point out though, hosting costs are too expensive to even consider trying to sell content at a reasonable price for the consumers.

    We also have to look at the current markets where broad band is the most available. Japan having FTTH to more customers would probably be the best market to start introducing better services. The problem then becomes a mentality problem over a technical or even economic problem. The Japanese are very low risk oriented. They won’t actually invest money unless they know they can get their investment back. Producing high quality content as well as hosting high quality content would take an astounding amount of capital which won’t be an easy sell. I don’t know Korean culture as well, so I can’t really say the same about the mindset that they have.

    I can say this about the experience of the broadband connction though. Having 100 mbps to home is nice, even if you can’t fully utilize it. Much like buying an SUV with 500HP and an off-road package while living in a San Jose Condo. I will probably never use near the full potential of the vehicle, but it is nice to know that it is available.

  • George Ou (author) said:

    Great points Michael. The Crysis analogy is good, but the application market is even more stingy about pushing the limits. Here we have over 40 million broadband customers as of 2008 who are capable of more than 6 Mbps, yet they’re being ignored. If there were just 10 million potential customers running faster graphics adapters and CPUs, a lot more game companies would be producing more advanced games.

    As it turns out, ad driven content is stuck in the 1 to 3 Mbps range while pay content is stuck in the 1 to 6 Mbps range. The cost of hosting is extremely prohibitive for HD video relative to the cost of residential broadband. The fundamental reason for that is that broadband is shared at 20:1 to 40:1 while hosting bandwidth is shared at 1:1 to 4:1. Furthermore, the content provider has to bear the cost of the total sum of all of their active users. That by definition makes HD video hosting very expensive.

    You’re also right that having a 100 Mbps connection is like having that 500 HP car. You’ll never be able to drive it to its full limit, but it’s nice to have the acceleration from 0 to 65. There are people who will pay for this kind of service, just not too many of them.

Leave your response!

Add your comment below, or trackback from your own site. You can also subscribe to these comments via RSS.

Be nice. Keep it clean. Stay on topic. No spam.

You can use these tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

This is a Gravatar-enabled weblog. To get your own globally-recognized-avatar, please register at Gravatar.