Posted: 09th Mar, 2009 By: MarkJ
The
Broadband Stakeholders Group (BSG), a UK government advisory think-tank on broadband, has posted an interesting update about how telephone line lengths affect broadband speeds. One area of particular interest is the scatter graph below, which plots line length versus average throughput speed:
(Left Side: Mbps | Bottom Row: Metres)
The BSG Said: "
Although using straight line length (the straight line distance from a home to the exchange) rather than the actual line length, the level of variance in performance between lines of comparable length is pronounced.
Getting behind the reasons for this variance should be a central concern of policymakers and the industry alike. The causes of the variance could have important implications for the development of public and regulatory policymaking in this area. We wait to see what Ofcoms second report on broadband speeds is able to say on this."
Admittedly this is nothing new and nearly all UK ISPs warn customers about the serious effect that telephone line length can have on broadband performance, which is part of how most speed estimates are created. The further you are away from your exchange the slower the service.
Never the less it's interesting to see such information visualised, thanks to data from
Samknows, though it's a shame the test didn't include actual line length because straight line length estimation can be highly unreliable.
One person we know lives just 1km from their exchange, but the wire loops around his area and ends up being about 7km in length. The impact on speed means that while a straight line length may predict performance in the high Mbps field; in reality he can only receive something nearer to 1Mbps.