Time Warner speed test.
If you’re an average US broadband user, you’ve had internet access for 10 years, spend about an hour a day online at home, and are enjoying far less of the bandwidth your service provider promises you. These unsurprising stats come from a new report from the Federal Communications Commission entitled (equally unsurprisingly in plain-vanilla bureaucratese) “Broadband Performance”, that provides detail on US fixed — not mobile — broadband.
There’s a wonderful irony here that the recent Google-Verizon suggested diminution of the FCC’s regulatory role also makes a clear distinction between fixed (to be regulated) and mobile (to be unregulated) broadband. But we digress…
The 30-page “Technical Paper No.4” was produced by the Omnibus Broadband Initiative, better known as the OBI, a core component of the FCC’s National Broadband Plan, in support of the commission’s efforts to address what it calls “the broadband availability gap”.
The majority of the report’s findings are typical of the need-to-know-but-could-have-guessed data nuggets that planners require to underpin any major government initiative. The interesting info, as is usually the case in such studies, is in the details.
Take, for example, the finding that actual broadband speeds are lower than as advertised by carriers. Amazing! Whodathunkit? Next the feds will be telling us that there is no Santa Claus, storks brings no babies, and that good beer tastes better when it’s not ice cold.
What is interesting, however, is the high degree of variability in actual versus advertised data-download rates, as determined by data from Akamai, comScore, and the FCC’s own research:
… in 2009, average (mean) and median advertised download speeds were 7-8 Mbps, across technologies. However, FCC analysis shows that the median actual speed consumers experienced in the first half of 2009 was roughly 3Mbps, while the average (mean) actual speed was approximately 4Mbps. Therefore actual download speeds experienced by US consumers appear to lag advertised speeds by roughly 50 per cent.
In its overview of possible causes for this discrepancy, the FCC mentions network congestion, “under-functioning wired and wireless home routers”, latency, jitter, and packet loss. Perhaps unsurprisingly, knowing the delicate negotiations in which the FCC is currenly embroiled, “bullshitting service providers” is not mentioned as one possibile reason for service not matching promise.
Another “Well, duh!” FCC finding was that some users consume far more bandwidth than do others. What was surprising, though, was how great the discrepancies could be:
“The extreme difference between average and median data usage is principally due to a relatively small number of users who consume very large amounts of data each month — sometimes terabytes…” the FCC explains. Furthermore, the data-suckiest one per cent of residential users “appear to account for” roughly 25 per cent of all traffic, and the top three per cent eat up 40 per cent.
If Cisco’s dream comes true and we’re all using the internet for high-def telepresence in some hyper-wired future, the median will move ever-closer to an ever-increasing mean, and Verizon’s FiOS/FTTP gamble will look mighty smart.