Average U.S. Broadband Connection Now 6.7 Mbps

Google Fiber will skew this data significantly. Looking forward to the report that includes this data.

doubtful since KC makes up less than 1% of the US.

I'd rather have a 6.7Mbps connection that is 20-30ms away from the server than a 15Mbps connection that is 100ms or more away, any day.

It is how you use it.

Yeah, as a country, we are that guy.

Latency 100 vs 30 ms means nothing to video streaming, browsing, etc. Gaming users and audio & video conferencing users care.

It's actually 0.19% of the US, so yeah.

I wish I could get half that or even a third as the best that is available in my area less than 10 miles outside a major city is Verizon DSL at 1Mbps. Thank you US government for allowing Verizon to have a monopoly to give me such poor speed for $24/month.

Laugh, but there is a kernel of truth to what you say. Once you get enough bandwidth (5 to 6mbit/s) to stream HD video the rest is purely gravy.

I was lucky enough to qualify for the 10 - 15mbit/s DSL package from Verizon before I moved in with my fiancee. I had 10mbit/s and change of bandwidth 24/7. At the new location my choices were 3mbit/s residential DSL, 5mbit/s cable, or 6mbit/s business DSL, I opted for the third option.

The loss of 4mbit/s sucks on paper but it hasn't really changed my internet experience in any quantifiable way. I can still stream HD videos. Webpages load just as snappy as they did before. My gaming hasn't been impacted in spite of the speed drop and change from fast path to interleaved. Large downloads would seem to take more time on paper but rare was the mirror site that I could find that would peg my 10mbit/s connection. Well seeded bittorrents could do it easily but outside of that I never found too many ftp/http mirrors that would peg the connection. Even CDN downloads like Windows Update usually top out at 3-4mbit/s in my experience, though YMMV of course.

Wouldn't that be what the people that they are tracking subscribe to? If that's the case then this is not an accurate representation of anything.

DSL is getting left in the dust. Telco's need to start writing state reg's to ban cable upgrades in addition to muni-broadband competition.

The only way to get back on the exponential growth wagon so we can keep up with silicon valley is with the help of FTTH.

Which if you think about it is very sad for anyone that has FIOS. Since they have no excuse for having speeds that haven't increased that much since it was first introduced.

If you add the geographic land of the first 10 of all these countries they are probably no bigger than less than 10 states.

The US infrastructure cost would be 5-7 times higher than ALL of those combined.

Japan is only about the land mass of California.

That is why these reports are bogus, because let's face it the urban centers have great connectivity, the rural not so great.

And if the US wants to piss away USF which could have built fiber to every home many times over, then you have a story.

You can stream HD with a reliable 2-3 Mbps, so even if you are this rate you can have passable streaming access.

I have 25/25 because I need the upload for my work, and even I could get by w/ the 15/5 from Verizon without any real impact to my work, except uploads would take 5x longer. Since this is mostly background, who cares.

I remember back in the dial up days it may take 1-2 days to download a file a few meg in size. You started the download, and went and did something else. Yes this is not as productive but I would hazzard that over 6 MBps (with good ping) that can handle multiple streams today, so while average it satisfies 95% or more of the population.

Where the big fail and nobody talks about it is UPLOAD. If upload rates remain patethic, then this will reduce the utility of cloud-based apps that need to exchange data upstream. For instance, this may delay remote desktops or other innovations, cloud backup, etc...

If you add the geographic land of the first 10 of all these countries they are probably no bigger than less than 10 states.

The US infrastructure cost would be 5-7 times higher than ALL of those combined.

Japan is only about the land mass of California.

That is why these reports are bogus, because let's face it the urban centers have great connectivity, the rural not so great.

And if the US wants to piss away USF which could have built fiber to every home many times over, then you have a story.

You can stream HD with a reliable 2-3 Mbps, so even if you are this rate you can have passable streaming access.

I have 25/25 because I need the upload for my work, and even I could get by w/ the 15/5 from Verizon without any real impact to my work, except uploads would take 5x longer. Since this is mostly background, who cares.

I remember back in the dial up days it may take 1-2 days to download a file a few meg in size. You started the download, and went and did something else. Yes this is not as productive but I would hazzard that over 6 MBps (with good ping) that can handle multiple streams today, so while average it satisfies 95% or more of the population.

Where the big fail and nobody talks about it is UPLOAD. If upload rates remain pathetic, then this will reduce the utility of cloud-based apps that need to exchange data upstream. For instance, this may delay remote desktops or other innovations, cloud backup, etc...

Read up on outliers and how they skew the mean. Even with only a small portion of the population, 1000Mbps in KC area will bump the mean about 2Mbps -- 30% spike.

what this report fails to take into account is latency and lost packet rates. docsis3 is faster, but just as dirty and noisy as the narrow band network channels it's binding together. latency may not be important while watching 720i youtube, but it's critical when fragging an opponent inside the latest mmo. ironically a lot of those Asian country networks have a wireless last mile, which often means they are faster, but their latency is far worse than cable modem coax wire based isps here in the USA. so while speed alone may be a useful marketing number any idiot can compare services to, it's far from telling the entire story of how well your isp is performing. it's kind of like buying your sweetheart a 3 caret diamond engagement ring, while ignoring all the carbon flakes and air bubbles floating around inside the stone.

Verizon does not have a state-sponsored monopoly on your service.
If you and your neighbors were willing to pay the freight, you would have a competing service available. But you aren't.

So you'll have to wait for LTE.

Let's assume EVERY person in KC (both cities) has 1gbps. That's roughly 610,000 people. There's 311,800,000 people in the USA. Math was never my strong point, but wouldn't you multiply 6.7 x 311190000 (2084973000), then 1000 x 610000 (610000000), add them together (2694973000), then divide by 311800000? So we'd be looking at... 8.6mbps average. A whopping 1.9mbps increase. Sure, it's a spike, but a tiny one (it doesn't change the data "significantly", and it's useless anyways because KC's combined population is only 0.19% of the US population).

There might be some holes in this because I should be using household figures instead of population figures, although I guess it's the same either way... I'm not sure.

It only shows what people are BUYING, not what is available. Those are two distinct things and the latter is hard to measure.

The quick fix would be to outlaw DSL and kick those subs to the curb. Like the unemployment rate, don't count them and it improves the average.

And within that 0.19%, less than a third will actually pay the $70 a month for the 1Gbit tier. Most will take the "free" 5Mbit tier, while plenty more will flock to TWC or AT&T, where double/triple-play TV bundles *with ESPN* will be half of Google's price.

... »www.akamai.com/stateofth ··· nternet/
and visually compare North America VS the rest of the world on two metrics average CONNECTION speed, and average PEAK speed.

When you consider the area and populations involved the US looks in pretty good shape, and rapidly improving.

You actually need to look at the number of households instead of the number of people. Each individual in a household would not have a separate subscription to Google fiber.

.

I just got 23/22 to New Jersey and immediately after got 3/2 to California. I'm paying for 15/5. These tests are just random number generators.

In the past we expected to be better then the rest of the world. We had the drive and desire to do it. But now we will always be "just average" with everything from now on. The people of the US have been programmed to accept "average." We don't seem to care any more. (You can debate how) With few exceptions, there isn't much desire from companies or consumers to deliver or expect anything more then average. (More cost $$$$. Average cost less $$$$.)

Akamai's State of the Internet Report has been and will always be bull shit. Why? Because it only accounts for what people SUBSCRIBE to not what the COULD subscribe to.

It won't make a dent anymore than any other FTTH provider that offers 1Gbps has. Remember Chattanooga has had 1Gb for at least a year and it's done nothing for this BS report.

I know this will be tough news for many here to take but the over whelming majority of users in Kansas City are going to take the almost free 5/1 connection over the $70/mo 1Gig because for most people 5/1 will meet their need and they are going to look at price over performance.

Yeah. I'd hate to see where the US would place should the report include what the people in S. Korea and Japan COULD get if they would pony up and buy the fastest tier offered. We'd really be wanting our 12th place back!

/M

yet another failure for telecom to act in the public interest

Maybe. Or maybe Verizon has already used their large $$ monopoly and bought off his states legislators and taken that right away. Don't worry the cable monopolies helped them do it.

I wish I could get 6Mb. I am stuck on the 1.5 Frontier DSL deal.