32 thoughts on “As Broadband Growth Slows, Expect Speed Boosts”

  1. I know this is off topic, but your chart makes me feel really good about my community’s cable-based broadband. We get 15M/768k for $45/mo. I’d get a discount if I were a senior citizen! I think that’s pretty amazing for an in dependent cable company with about 4000 Internet subscribers that’s playing in the same market as Comcast, fios, and anyone who can put their dslam in our local CO.

    I guess the relevant question is how fast can the carriers push their systems to without upgrading end user equipment.

  2. I hope they offer more upstream as well. More and more folks will want to back up files off site, and send video and high quality audio, and we’ll need good 1.5mbit/sec upload or better for that, and the technology can do it.

  3. Time Warner is offering RoadRunner Turbo with 15Mbps down and 1Mbps up. You may want to update the chart. I don’t know what the price is though – I haven’t seen a need to upgrade.

  4. Higher speeds will also be mandatory for those that are moving their computing infrastructure to the cloud. Waiting for downloads & uploads simply will have to high an impact on productivity. I would have FIOS today if it was available in my “neck of the woods.” That said, we are planning a move and FIOS availability, while not the driving factor, is a considerations.

  5. Om,

    Isn’t it time we dispensed with the misconception that “peak downstream speeds” are relevant? All providers oversubscribe bandwidth (and, in the case of cable operators, clamp down on bandwidth if you actually hit the peak (and, no, this isn’t the same as what Comcast was doing to P2P)). They do so for perfectly logical reasons and this isn’t necessarily an evil thing. But as more and more traffic is made up of over-the-top, long-form video, peak speeds mean nothing. Average speeds are all that matter and they are often as little as one-third to one-tenth the peak speed.

    Unless providers develop the means to ship video at near-peak rates, you might do your readers a service by trying to ferret out what the average rates actually are.

  6. @ Kevin Walsh,

    Excellent point. And I agree. As I said in the last paragraph, there are multiple factors that control the delivery of content, and as a result we shouldn’t get too excited by top speeds. This is all “marketing” spin. I want SLA and QoS from my access provider.

  7. whatever happened to all the bonded copper companies that were around? last mile nirvana is always a few years away…

  8. ha, india, as a consumer, i can get about 80 kbps, max … never saw a whole youtube video from beginning to end

    the web will be like luggage when you travel, your stuff will expand to fit the available capacity, i actually think it will soon be slowing down even in the us …. hd tv, every kind of online service and application, streaming everything… we shall see

  9. As a US resident, and a New Yorker I have to say, FINALLY! I’m paying for what is SUPPOSED to be 10Mb/sec internet access and getting MAYBE 2. I’m a Cablevision customer and the service has been nothing less than appauling. Even Youtube videos don’t stream without buffering in the middle. It’s sad really. Just last week or so we finally got our first real competitor in the area with Verizon’s FIOS service, If it gets any worse at Cablevision, I’m jumping ship.

    Now that broadband is becoming widespread throughout America, there needs to be a bigger push for building bigger and better infrastructure with more bandwidth for consumers. We are falling FAR behind most of the developed world when it comes to internet speeds, especially when you look at countries like Japan. If the push in the tech sector is to put software in the “cloud” and have the computer, or cellphone, or device in general become just a general purpose terminal for computing with web applications, we are going to need serious increases in bandwidth and speed.

    I’m not waiting a half hour for an online photoshop to load, I’ll just use the real one.

  10. Kevin: that’s the critical point, especially if you’re talking about broadcast 2.0 where people are increasingly replacing TV viewing with streaming content. I don’t think people realize just how big the discrepancy between peak and average rates are. With DSL, a common over-subscription ratio is 20:1 (i.e. if all subscribers started downloading a file simultaneously, they’d all get 1/20th of the advertised peak rate). And let’s not even talk about the fact that the peak DSL rate drops off significantly with distance from the DSLAM. Cable is just as bad. Typically, that 150 Mbps will be shared between 500 and 1000 subscribers (though the cable guys can improve things by making more channels available for DOCSIS). Right now, FIOS and uverse are the only systems that can give multi-megabit/sec average rates.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.