FCC Relying On Faulty ISP Performance Data

alphadogg writes "The FCC recently used speed test results from comScore as an absolute indicator of specific ISPs' performance. Consulting firm NetForecast analyzed comScore's testing methodology and data to assess whether it accurately reflects broadband ISP performance, and to assess the appropriateness of using the data to reach general conclusions about the actual performance ISPs deliver to their subscribers. NetForecast uncovered problems on both counts. They found that the effective service speeds comScore reports are low by a large margin (PDF) because its data calculations under-report performance and place many subscribers in a higher performance tier than they purchased."

Comments

FCC is faulty?

[Bob_Who]

I just am so surprised. Its run by a bunch of government employees, and they are rarely faulty.

Re:FCC is faulty?

[Otterley]

And we all know employees of private companies are infallible.

Re:FCC is faulty?

[Lunix Nutcase]

No, but in contrast to a government operation there are a dozen other companies colluding with each other and trying to screw their consumers as much as possible.

FTFY.

Re:FCC is faulty?

[Tubal-Cain]

US Code Title 47, Chapter 5, Subchapter I, Section 151:

For the purpose of regulating interstate and foreign commerce in communication by wire and radio so as to make available, so far as possible, to all the people of the United States, without discrimination on the basis of race, color, religion, national origin, or sex, a rapid, efficient, Nation-wide, and world-wide wire and radio communication service with adequate facilities at reasonable charges, for the purpose of the national defense, for the purpose of promoting safety of life and property through the use of wire and radio communications, and for the purpose of securing a more effective execution of this policy by centralizing authority heretofore granted by law to several agencies and by granting additional authority with respect to interstate and foreign commerce in wire and radio communication, there is created a commission to be known as the “Federal Communications Commission”, which shall be constituted as hereinafter provided, and which shall execute and enforce the provisions of this chapter.

comScore got it more or less right

[Spazmania]

comScore got the data more or less right. The OP's main complaint seems to be that the speed is under-reported because packet loss causes the TCP session they used to slow down. Guess what? Packet loss causes the TCP session to slow down. Customers on ISPs with noticeable loss rates experience slower performance than the line's rated speed. Hello!

Re:comScore got it more or less right

[COMON$]

Ya I pay for the "Extreme" Roadrunner in my area. which gives me a better upstream for my telecommuting wife. supposedly 10M/1M but it is more like 3M/768k, most of this is due to really high latency and dropped packets. When it works it works, so I guess by this guy's definition I get my 10/1, just as long as you don't count the packet loss...

Re:Wait for ACK?

[topham]

TCP/IP doesn't wait for the ACK. It keeps sending until the Window is full, or the ACK is received. If the Window fills it will wait until the ACK is received (or timeout and retry, etc).

If the test is trying to automatically place the users in specific Tiers then there could be a problem, however the rest of the issues are mostly a red herring. I use Speedtest.net and can readily attest to it's general accuracy, and I seriously doubt any other services are all that different.

by the way, I'm not in the U.S., I actually get what I pay for.

Re:Wait for ACK?

[Shakrai]

by the way, I'm not in the U.S., I actually get what I pay for.

You might have worded that a little bit better. Canada and Australia have worse broadband networks than the US does. Most US users on DSL get what they pay for. Cable networks may or may not deliver the promised performance at all hours, but that's simply the nature of the beast. In my area Time Warner provides 10MBit/s service on a DOCSIS 1.1 network. That means that just four customers are enough to max out a node that serves dozens to hundreds of customers.