dslreports logo
 story category
Study: U.S. Fifteenth In Broadband Quality
Having some kind of broadband plan is apparently a good idea.

A new study by the University of Oxford in England and the University of Oviedo in Spain measured broadband speed and latency in some 66 countries and 240 cities, ranking countries on their ability to handle next-generation broadband applications and services and the quality of a nation's broadband connections. Mirroring the results of most other studies, South Korea and Japan take the top two spots, while the United States ranks fifteenth behind countries such as Lithuania, Denmark and the Netherlands.

Click for full size
The report claims top ranking countries saw "increasing investments in fiber and cable network upgrades, coupled with competition diversity, and supported by strong government vision and policy." Given many markets lack two or three of those, the country's ranking is once again not particularly surprising.

Sure, the United States is larger than many of these other countries, and providing broadband to rural customers is an obvious challenge. But the United States also fails to break the top twenty best cities for broadband, five of which are in Japan. You'd think there'd be at least one U.S. city able to bump either Vilnius, Lithuania or Kocise, Sweden out of the top twenty.

On the positive side, the United States has made some "significant, above average improvements" in infrastructure, according to the researchers. It's an obvious assumption that much of this improvement is courtesy of Verizon and Comcast's investment in fiber to the home and DOCSIS 3.0, respectively.

"The Broadband Quality Study shows us which countries have made real moves towards the Internet of the future," says Professor María Rosalía Vicente from the University of Oviedo. Of course as most of you are aware, the FCC is cooking up the United States' first ever national broadband plan, which will be unveiled in exactly 139 days.
view:
topics flat nest 

Matt3
All noise, no signal.
Premium Member
join:2003-07-20
Jamestown, NC

Matt3

Premium Member

Anyone notice two data points?

Did anyone notice that they not only track upload speed, but latency as well? I find it sad that the global average upload speed is 1.3Mbps, but Time Warner only provides 384Kbps or 512Kbps in my area. How pathetic.

MoJeeper
The Stig in 2012
Premium Member
join:2000-10-20
Springfield, MO

MoJeeper

Premium Member

Re: Anyone notice two data points?

Your lucky AT&T can't even get me a stable dialtone.

ArrayList
DevOps
Premium Member
join:2005-03-19
Mullica Hill, NJ

ArrayList

Premium Member

Re: Anyone notice two data points?

thats pretty bad considering that AT&T has been in the dialtone business since, well forever.

hobgoblin
Sortof Agoblin
Premium Member
join:2001-11-25
Orchard Park, NY

hobgoblin to Matt3

Premium Member

to Matt3
said by Matt3:

Did anyone notice that they not only track upload speed, but latency as well? I find it sad that the global average upload speed is 1.3Mbps, but Time Warner only provides 384Kbps or 512Kbps in my area. How pathetic.
2mb here.

Hob

Matt3
All noise, no signal.
Premium Member
join:2003-07-20
Jamestown, NC

Matt3

Premium Member

Re: Anyone notice two data points?

said by hobgoblin:
said by Matt3:

Did anyone notice that they not only track upload speed, but latency as well? I find it sad that the global average upload speed is 1.3Mbps, but Time Warner only provides 384Kbps or 512Kbps in my area. How pathetic.
2mb here.

Hob
Fortunately for you, you're not in my Time Warner area.

FFH5
Premium Member
join:2002-03-03
Tavistock NJ

FFH5 to Matt3

Premium Member

to Matt3
said by Matt3:

Did anyone notice that they not only track upload speed, but latency as well? I find it sad that the global average upload speed is 1.3Mbps, but Time Warner only provides 384Kbps or 512Kbps in my area. How pathetic.
You must live in wrong state.




Matt3
All noise, no signal.
Premium Member
join:2003-07-20
Jamestown, NC

1 edit

Matt3

Premium Member

Re: Anyone notice two data points?


Me
I have 15Mbps/2Mbps from my local ILEC, delivered via GPON fiber. See that Jamestown result in #10 for North Carolina? That's my connection skewing the result.

PapaMidnight
join:2009-01-13
Baltimore, MD

PapaMidnight to FFH5

Member

to FFH5
Rhode Island, on paper, looks like the place to be. But beware of those Speedtest averages. PowerBoost inflates upload speeds on paper. In real world environments, they can be 2/5 the speed, or, in my case, as much as 1/5 (I've been powerboosted to 10mbits up... )..

Matt3
All noise, no signal.
Premium Member
join:2003-07-20
Jamestown, NC

Matt3

Premium Member

Re: Anyone notice two data points?

It also looks like they accept business and educational connections, which makes the ranking of 15th even more pathetic, because residential connections on average will be MUCH lower.

FFH5
Premium Member
join:2002-03-03
Tavistock NJ

2 edits

FFH5

Premium Member

Re: Anyone notice two data points?

said by Matt3:

It also looks like they accept business and educational connections, which makes the ranking of 15th even more pathetic, because residential connections on average will be MUCH lower.
That data can be easily excluded because the tests clearly identify when a test is from a company or school. Did the universities that did the study exclude that data? Who knows, without pouring thru their whole report. But the key would be were they consistent across all the countries.
WhatNow
Premium Member
join:2009-05-06
Charlotte, NC

WhatNow

Premium Member

Re: Anyone notice two data points?

Pembroke the town is not much bigger then the University and if you are not in one of the apartments that border it your speeds are hit and miss due to where RTs sit and the equipment installed in them. It goes from town to large farm fields within 1/2 mile of the downtown / University area. If Walmart and a very big hardware store that caters to farm supplies meets your shopping needs then you would be very happy to move there. But it would pay you to check what the highest speed internet you can get at that address. It may be dialup or just 3meg dsl.

South Korea
38,622 sq. miles
48,379,392 population
1,274/sq mile pop density
vs
Pennsylvania
46,055 sq. miles
12,448,279 population
274.02/sq. miles

10th in population of US states

I bet if Pennsylvania added 1k of pop/sq. miles the ISPs could up their speed. It costs a lot of money per foot of any kind of cable copper or fiber. If you have a high density your costs per customer is a lot less and your ROI goes way up. If you just took the internet speeds

I would love to see FTTP to everybody that is on the power grid but unless the ISPs start charging a lot more from the existing customers to pay for for the fiber don't look for it. I predict the customers that already get a decent speed will get it before the semi-rural areas outside Pembroke, NC does.
Expand your moderator at work

FFH5
Premium Member
join:2002-03-03
Tavistock NJ

2 edits

FFH5

Premium Member

Study based on www.speedtest.net data

The study was based on data gleaned from »www.speedtest.net/global.php web site where you can do your own drill down analysis.

How good and comprehensive this data is can be debated, since the collected data is a self-selection process and only reflects those who have gone to the site to do a speed test. That isn't generally a valid way to do statistical analysis, though some verifiable data is better than none.

If you can come up with the scratch, you can submit a query here to get your own customized reports:
»www.speedtest.net/datain ··· uiry.php

»www.sbs.ox.ac.uk/news/me ··· 2009.htm
Using more than 24 million records from actual broadband speed tests conducted by users around the world in May 2008 and from May to July 2009 through www.speedtest.net, the research team calculated statistical averages for each country of several key performance parameters used to determine the quality of a broadband connection.

These parameters were grouped into three major categories: download and upload throughput, and latency. The Broadband Quality Score (BQS) for each country was determined using a formula that weighted each category according to the quality requirements of a set of popular applications now and in the future.

Typical applications for today include web browsing, social networking, music downloads, basic video streaming and video chatting, standard definition IPTV, and enterprise-class home offices.

Future applications include consumer telepresence for communications, healthcare and education, high-quality video file sharing and streaming, high-definition IPTV, cinema-quality live event broadcasts and advanced home automation.
FFH5

FFH5

Premium Member

US vs EU not far apart

If you look at the United States and compare the numbers to the EU and then treat separate US states the same as separate countries in the EU, there is a close comparison:




jmn1207
Premium Member
join:2000-07-19
Sterling, VA

1 edit

1 recommendation

jmn1207

Premium Member

Re: US vs EU not far apart

I wonder if the total number of broadband users has any impact on the quality? Other than China, which is surprisingly low on the total number of broadband subscribers by percentage of their population, no other country in the world comes close to the US in the total number of broadband users.

How many different ISP's are available in the Netherlands, Sweden, or Denmark? We probably have more DOCSIS 3.0 Comcast subscribers than all of those countries' broadband users combined. If we were to use just FiOS and Comcast as our benchmarking results, we would be doing much better.

By kicking out the bottom half of every nation's broadband users, I bet most of these other top performing countries would see nearly the same stats. However, I bet the US would skyrocket upwards if we were to only use the best 40 million internet connections. We simply have more choices and a much wider variance from top to bottom. In other words, the first 2.5 million Dutch broadband users are probably nearly identical in performance to the rest of the 2.5 million or so users. That is not likely here in the US, where surely the top 40 million enjoy much better speeds and latency than the bottom 40 million "broadband" users.

»en.wikipedia.org/wiki/Li ··· et_users

Boogeyman
Drive it like you stole it
Premium Member
join:2002-12-17
Wasilla, AK

Boogeyman

Premium Member

Huh?

How can two foreign uni's have a better (or any) broadband map when our own government doesnt know? Maybe instead of ConnectedAmerica or whatever the hell that outfit is called, we should hire these people to map our BB coverage.

FFH5
Premium Member
join:2002-03-03
Tavistock NJ

2 edits

FFH5

Premium Member

Re: Huh?

said by Boogeyman:

How can two foreign uni's have a better (or any) broadband map when our own government doesnt know? Maybe instead of ConnectedAmerica or whatever the hell that outfit is called, we should hire these people to map our BB coverage.
See »www.speedtest.net/global.php

You can drill down to cities in the US and see what ISPs service what areas. Any US company or gov't group like the FCC can get a detailed report from speedtest.net just like those foreign universities did.

houkouonchi
join:2002-07-22
Ontario, CA

houkouonchi

Member

Japan is #1

Japan is much higher quality broadband than Korea. Sure you can get fast speeds in Korea maybe but Korea has shitty routing to the U.S. and other countries were a lot of the servers are. Japan has excellent connectivity to the U.S and around the world. I have seen 80+ megabit speeds to my server in Los Angeles from a residential connection in Japan which is much better than you would ever get in Korea.

Ikyuao
join:2007-02-26
Wichita, KS

Ikyuao

Member

Re: Japan is #1

Exactly. That Japan really has higher quality and excellent connectivity of an internet right there.

DaveDude
No Fear
join:1999-09-01
New Jersey

DaveDude

Member

Wash, rinse Repeat.

These articles are like wash , rinse repeat. Really go back in history, and see there is one for last month, and the month after, etc ...
donh06
join:2006-11-27
Jackson, WY

donh06

Member

We must be winning at something.

I think we must be leading at something. Maybe we should be looking at different numbers. Maybe ceo pay, customer dissatisfaction and profit. I am sure we will come out better in those. Two of those are what drives decisions here so surprise they will be what we win at. Customer satisfaction is a small consideration. You just need to keep the competition out so nobody can steal your customers away.

Don

zpm
join:2009-03-23
Columbus, GA

zpm

Member

Re: We must be winning at something.

15th, well were also 15th in the poll of happiest countries in the world.

rjdriver
@cox.net

rjdriver to donh06

Anon

to donh06
Yeah, we're leading in how effectively our broadband suppliers ration out the bandwidth in the just the barest level needed to keep us from sceaming bloody murder about it.

DOCSIS 3 should have changed everything. Why didn't it? Lack of real competition and poor governement oversight and regulation(aka - corruption).
openbox9
Premium Member
join:2004-01-26
71144

openbox9

Premium Member

Re: We must be winning at something.

DOCSIS 3 has allowed cablecos to double or in some cases quadruple their throughput offerings. Seems like a change to me.

tmh
@qwest.net

tmh

Anon

I have to ask....

Did anybody study what kind of extortion tactics CAPS they have in place for heavy users?

doublea
join:2007-06-04
Rancho Cordova, CA

doublea

Member

It's all come about in the last 2 years.

I will say my area is much improved in broadband prennetration (north bay, California) in the last 6 months. We got uverse here 6 months ago, and comcast just finished rolling out it's docsis3 network. And our local ISP sonic.net is working hard to get it's 24mbps service out to as many people in our area as possible.

Att finished putting in a make shift 3g network only 10 months ago. Spring and verizon have both had 3g here for 1.5 years.

A look back in time in the beginning of 2008 we had no 3g, no docsis 3, and no vdsl. It has all poped up here in the last 2 years. It was a grim outlook then.

joebarnhart
Paxio evangelist
join:2005-12-15
Santa Clara, CA

joebarnhart

Member

Small providers kick @ss

Click for full size
WAN quality, packet loss and latency for my ISP
Find the small providers in your area. I have Paxio, which is probably the smallest ISP in the S.F. Bay Area but it has the lowest latency and the highest speed connections. Don't just swallow the Komcast Kool-aid without checking to see if there's a better choice in your area. I did, and looky what I have now!

qworster
join:2001-11-25
Bryn Mawr, PA

qworster

Member

California's nowhere!

California, home of Silicon Valley doesn't show up anywhere.
owensdj
join:2001-09-05
Easley, SC

1 recommendation

owensdj

Member

Let Government Build Fiber

The only way we're going to have broadband as good as those other nations is to have government build fiber optic to every home and business. Let multiple ISPs compete for service over the government-owned fiber.

XBL2009
------
join:2001-01-03
Chicago, IL

XBL2009

Member

Re: Let Government Build Fiber

said by owensdj:

The only way we're going to have broadband as good as those other nations is to have government build fiber optic to every home and business. Let multiple ISPs compete for service over the government-owned fiber.
Allowing private corps to create regional monopolies and own the last mile is a mistake.
owensdj
join:2001-09-05
Easley, SC

owensdj

Member

Re: Let Government Build Fiber

Huh? The government would own the last mile and work with many private ISPs to provide Internet service.

XBL2009
------
join:2001-01-03
Chicago, IL

1 recommendation

XBL2009

Member

Re: Let Government Build Fiber

We all own the roads and get to drive on them. Why should the roads of the internet be own by private corporations that only want to put up tollbooths and speed bumps everywhere.

AccessDelray
@bellsouth.net

AccessDelray

Anon

A National Broadband Plan

Using Broadband Infrastructure to promote Jobs Creation, Workforce Development and Entrepreneurship Education through the building of Social Entrepreneurship and Asset-Based Community Development modalities.

»www.digitalcommunitiesbl ··· _nation/
bsoft
join:2004-03-28
Boulder, CO

bsoft

Member

Fail

So, we're ahead of Germany, France, Italy, and Canada? And apparently our 2009 quality score is similar to Korea's 2008 quality score?

Our score improved dramatically in the last year because Comcast has been pushing DOCSIS 3 and Verizon has been pushing FiOS.

The reality is that the US could 'rank up' significantly in these types of rankings simply by lying about how much bandwidth we have. Provision 100Mbps DOCSIS 3.0 for every cable subscriber, and suddenly we look great. The HFC networks can't really deliver 100Mbps for each customer, and the providers have no incentive to offer it for a low price, but it's doable with toady's technology and with a minimal amount of investment.

The myth is that we need to be able to deliver 100Mbps with reasonable contention ratios to be able to have Korean-class broadband. The reality is that the contention ratios in Korea are anything but reasonable, and that the service can only deliver 100Mbps because relatively few people actually use it.