Measuring Broadband America Report Released
AzTechGuy writes "Early this year I received one of the 'Whitebox' routers to test the speed of my ISP and compare it to the advertised speed. Today I received an email that they have released the first report with another report due at the end of the year. My results do not correspond with the results reflected in the report."
It appears that most ISPs are within 80% of their advertised speeds during peak hours with Verizon leading the pack mostly exceeding their advertised rates. Cablevision users, on the other hand, shouldn't expect more than half of the promised bandwidth (youch!).
FIOS is awesome, and everything else is less than awesome. Pretty good advertising material for Verizon.
You mean a single data point doesn't follow the trend? Throw the study out! It must be crap!
Your hair look like poop, Bob! - Wanker.
I would say that any company that was listed that never reached an average of their advertised speed should be taken to task for false advertising. Maybe dragged in front of the FTC, and possible legal action since it looks like most providers are never able to deliver their advertised speed.
Time to offend someone
Cablevision users, on the other hand, shouldn't expect more than half of the promised bandwidth (youch!).
"Promised bandwidth"? I'm sure if you read the fine print on *any* residential broadband SLA, you'll find the ISP "promises" exactly *zero* bandwidth. Every contract I've ever seen says they promise speeds "up to" a certain amount but there is no lower limit to what they actually deliver. This is akin to the good old days of zero CIR frame relay where the provider had the right to discard up to 100% of your packets if network congestion became an issue. In return, you got rock-bottom pricing. I never saw any ISP ever drop 100% of packets due to a zero CIR, so you were essentially gambling (and winning) that the ISP would always have some spare bandwidth.
I'm sure people who opt for the 22Mbit/sec package expect they should get 22Mbit/sec or something close to it on a regular basis, and if the ISP is only regularly providing, say, 2Mbit/sec then the customer has a reason to be upset. However, to say the ISP is "promising" bandwidth is a complete fabrication. The OP should read up and understand the different between "up to" and "no less than."
In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky
Most consumer Internet does not promise any guaranteed rate or speed, only that it can peak "up to" a certain speed.
If you want guaranteed performance, you will probably have to pay for a business line, which is far more expensive.
"Two things are infinite: the universe and human stupidity; and I'm not sure about the universe." Albert Einstein
You know FCC could you please add hyperlinks to your PDF so I can easily go to the desired section?
TFA - Verizon: Sure, you can burst above our advertised rates, but enjoy that bandwidth cap!
I wonder how long this will last until a class action suit.
I think the ISPs are hiding behind the variables like distance to the tap and peak hours to not make a good faith effort to provide what they are advertising.
In many cases people pay for 3mbs but get 2mbs, then upgrade to the 6mbs plan and get 4mbs, which demonstrates the ISPs capability to have delivered the full 3mbs in the first place.
http://en.wikipedia.org/wiki/Jury_nullification
FIOS is dramatically outperforming even its impressive advertised speeds. This means that teenagers can infringe content and be lured by online predators that much faster! Quick, parents, sue Verizon for false advertising! You are getting too much (dangerous) bandwidth!
Cablevision users, on the other hand, shouldn't expect more than half of the promised bandwidth
Hmm...Perhaps you may not hit the max advertised rate on Cablevision's Optimum service, but I can tell from experience that it is much faster than most other services. I have Time Warner Road Runner Turbo and I am paying $66/mo for it in Western NY. I MAX out @ 1.7 MBps sustained, with bursts up to 2.0 MBps (Yes, Mega BYTES, bot bits). But when I visit my friends who live in Eastern NY where Time Warner doesn't have a death gripping monopoly on the broadband market, they are paying far less per month for speeds that always exceed 2.0 MBps on STANDARD level service. Optimum Online Boost, which some do have, get in excess of 3.0 MBps.
So in my personal experience, Optimum wipes to floor with other ISPs. Especially because they have no enforced cap like Comcast or FIOS, and are faster than Time Warner and Cox based connections. Benchmarks and speed tests are fine, but my real world use will decide what ISP I look for when it comes time to buy a house somewhere else. I don't care if they only give me 1/10th of their advertised speed. As long as that speed is still faster than the competition for an equal or lesser price, which so far they have been delivering in my experience.
"I hope you know how very lucky you are to know me, because I am so incredibly incredible."
If these speeds are averages over several days, and we are looking at 24 hours of data, then why in the world does the right side of the graph not mirror the left side of the graph? Surely the speeds don't suddenly change at midnight?
For most countries with good broadband 100MBS up AND down is fairly standard.
Over the past year or so, I've noticed that my Steam downloads - which would usually cap out at 1.2MB/sec - have been topping out at around 330K/sec after I get home from work. Other people I know have had problems with video streams constantly buffering when watching live video. I've considered upgrading to their "Boost" service, to see if that would help, but if this graph is any indication, it won't matter one bit. I priced out Fios a few years ago, when they first rolled out in my area, but I think it may be time to compare the pricing again.
Makes me wonder what the increased prices in the cable bill are going towards, if they're not building out any network capacity at all.
FIOS is soo good because not a lot of people are on it!
Once FIOS expands, I'm sure the network will get bogged down and speeds will drop to below advertising speed.
FIOS is not available in my area. Managed to talk to a Verizon sales person and said that penetration is slow because of regulation and dominance of the larger cable companies.
I would kill for FIOS!!!
Averaging hides highs and lows. I'm more interested in percentage of the time that I am receiving the advertised rate. It's relevant for connectivity in certain apps. For downloads total % of advertised bandwidth is relevant, but if apps are timing me out and at other points in time I'm getting 300% of the advertised speed, that doesn't correspond to good performance.
This system would be easy for any ISP to game. QOS routing is already in place in all ISP networks. All any one of them would need is an example whitebox (eg. one of their employees or their friends), and they could ensure all packets destined for the target host are treated with the highest priority. All we can tell from that graph is CableVision doesn't do that...
Remeber too that ISPs route packets differently depending on the destination provider among other things. Anyone remember the debacle about Comcast refusing to peer Level 3? They thought the traffic was lopsided, and as a result, all Netflix customers on their network were routed over a congested transit link which they refused to embiggen.
I signed up for Comcast 20mbps then my modem died and bought a docsis 3.0 modem and now I get 62mbps. The maximum I get from a given site is usually 20mbps there is the rare sites where I can download at 50+, but is awesome for doing multiple things. I set the USENET download at 20mbps and I can still play online and make voip calls at the same time.
I was also a participant in the study. My DSL service before the study was very poor. About a week and half after I received and connected my 'Whitebox' router my service got significantly better. You can see the performance was better (fewer dropped packets and higher substained bandwidth) in the personal graphs I received from Samknows after the first week and half. I have a feeling the my provider detected I was in the survey and made sure my traffic was prioritized.
.
A box-whisker graph would give a much better sense of how customers are faring.
Moreover, the tests weren't run blind. The ISPs provided data to the people running the study to help them disambiguate whether bottlenecks were in the last mile or in delivery to the ISP. It wasn't clear to me why that was even needed as sluggishness from the cloud would be spread evenly across all ISPs unless an ISP was cramming 100,000 users onto a single T1 line.
I have to admit since I finally bit the bullet and went to their cable modem 20MB service, the thing virtually always test out at 31MB/s for looong periods of transfer, I have been quite happy. It's a little expensive, but it is fast, and (so far) consistently 30Mbps/5Mbps down and up - not just in short bursts (I was worried about that.)
Loading...
Around here Comcast is basically the same speed as Qwest is, the main difference is that they have that boosting technology that lets them borrow bandwidth from the loop when congestion is low.
Unfortunately, neither company is particularly interested in doing anything about it and Qwest has flat out stated that they won't be upgrading capacity in several Seattle neighborhoods leaving them with a connection of 1.5mbps max or comcast.
I agree, in my opinion cable TV and customer service are extremely suspect, but I have the bargain Comcast service and can reach 15-20mb/s consistently.
From the report: "The data in this Report is based on a statistically selected subsetof those consumers—approximately 6,800 individuals—and the measurements taken in their homes during March 2011."
Interestingly, after a year of complaining to my ISP, using the data from my SamKnows router, that I was not getting anywhere close to my advertised download speed, in March 2011 I suddenly started getting it ( an ~4x boost). I wonder if my speed will now start to degrade now that the report is out? Or is it just coincidence?
how do i switch from the comcast i have (that is nigh unusable most evenings) to the one they tested (which is nigh 100% all day)?
It sucks so bad... We now have the 18mbps (upgraded from 6 to 12, and now to 18) plan and I haven't gotten over 6mbps since upgrading. It is actually slower now than it was with the 6mbps plan. Usually during peak hours I'm seeing under 2mbps and can't even play games or watch youtube videos above 240p.
I ran/submitted Uverse to this FCC test almost a year ago, it was much faster then and we actually got speeds as advertised. Me thinks the ISPs might have a way to game these metrics. I mean, its the FCC, I would hope private industry could outsmart a bunch of federal bureaucrats if they can survive in business.
My broadband comes from the town utility system. We reliably get speeds 90% faster than we pay for, at any time of day, or night. So-called 'LUSFiber' (for' Lafayette Utility System Fiber') is the best there is. Their cable offerings are also excellent and beat the opposition (mostly Cox) hands down. What is really cool is that the LUSFiber system stays up, even during a power outage, which none of the opposition does. As we live in an area that gets hurricanes, this is an important advantage. I wish everyone could get their service, it would put the big players to shame.
I've been a consumer as part of the study (and thats why I'm submitting as anonymous).
I'd agree with AzTechGuy, my rates for my west coast cable provider don't seem to be listed even though I'm part of the study.
The funny thing is that the report accredits all the "participating" broadband companies for their help in the study. Funny, cause when I sighed up they said the companies were not involved, and did not know I'd be participating, but funnier still just after I signed up my performance got better...
Hey I get to keep the white box when the study is done...
How long it took the ISP's to detect the boxes and set their throttle algorithms to ignore streams from them.
Now that Baorakus Obamaus has vanquished his opponents the FCC (Communications Intellingincia) knocks on his Highest of Holies Door.
Short story. The FCC's numbers are Fraudlent as their methods as their "Appointed Knights" of the Relm.
Good to be NOT of their Relm. //--++
Just out of curiosity, what is your max bandwidth? You on a 12 or a 20 or a 40? The reason that I ask is that 12s are adsl , while the 20/40 are vdsl. We are switching from comcast to qwest due to the outrageous price by comcast and their outages. In addition, I had to fight for vdsl rather than the adsl.
I prefer the "u" in honour as it seems to be missing these days.
Employee here, & CableVision was way, Way, WAY FASTER than TimeWarner was in the time I was working for CableVision (decent enough shop too imo, decent bosses as well - surprising & refreshing) in my experience. This was in 2003 whilst I lived in NYC's Long Island area (only minutes from the NOC where I worked etc.). I was blown away in fact by the diff. in speed in TimeWarner "RoadRunner" in those days (which I thought was fast in fact) once I got on CableVision's network!
That's speaking as both a user, and an employee then (former) mind you... CableVision told me, when I left (suddenly, big family type crisis):
"Come back ANYTIME, & you'll have a job here!"
(Right after I solved a problem in DELL systems being set "too radically" in their NIC config not working with CableVision's routers/cablemodems - which DELL blamed US for, figures... so, that was nice to hear + know!)
* So IF I'm ever in NYC again? I probably will look them up once again! Good place to work too I have to say, again... others mileage MAY vary, but not mine... good shop!
APK
P.S.=> I liked what I read about Verizon though (DSL man here nowadays - for me, the "price is right" with the mixture of the speed I require: Nice to know I am truly getting my money's worth, & THEN some, per this article's summary above (wish I could get FIOS though here - I can't currently, as my location is a "wee bit odd" so-to-speak (I am in the inner city, but in a strangely isolated area in it, hard to explain)))...
... apk
P.S.=>
I've read reports of some ISPs not supplying business connections to customers on residentially zoned land. I've read other reports of ISP sales reps not knowing what to do when a customer wants TV service from the ISP's home division and Internet service from the ISP's business division.
In many cases people pay for 3mbs but get 2mbs, then upgrade to the 6mbs plan and get 4mbs, which demonstrates the ISPs capability to have delivered the full 3mbs in the first place.
Then perhaps what customers are paying for is a slice of spectrum x MHz wide, and the Mbps per MHz ratio depends on line conditions such that people living farther away will need to pay more for the same last mile service because it costs the telco more to provide the same last mile service.