Broadband bone of contention

WITH BROADBAND speeds increasing, Ireland appears to be catching up with other countries’ super-fast networks

WITH BROADBAND speeds increasing, Ireland appears to be catching up with other countries’ super-fast networks. However, the small print leaves many customers with far less than they pay for.

While most of the attention on broadband statistics is given to the headline speeds, the issue of line contention is now proving to be equally important.

Contention refers to the number of customer lines that share a connection in a local network exchange, with the top speed being split between them all. As a result, a customer on a 20MBit/sec broadband connection with a contention of 20:1 may end up with speeds as low as 1Mbit/sec during busier times as the top speed of 20Mbit/sec is in reality shared with 19 other customers.

Contention is one of the reasons why broadband speeds are generally advertised as “up to”, although the quality of the line and the customer’s distance from the exchange – where local connections are passed through to the wider network – are also a factor.

READ MORE

“Some telecommunications companies, because of a lack of capital or as a way of squeezing the customer, design in contention to their network so that a number of people will share an amount of bandwidth,” says Mark Kellet, chief executive of Magnet Networks, which offers uncontended lines so the customer gets access to all of the potential speed without having to share.

“So while companies like Eircom advertise 24Mbits/sec the chance of you getting that speed during the busy period is very slim because you’re sharing with other people.”

At present Eircom and Vodafone offer a contention ratio of 48:1 on their residential broadband packages, while cable company Chorus/NTL set theirs at 17:1.

Mobile broadband is also highly contended. However, exact figures differ depending on the network a user is on, the mast they are connected to and how many other connections are using it at that time.

A survey in 2008 by broadband communications company Epitiro found that Irish landline customers were getting just 60 per cent of speeds advertised by their suppliers, while mobile broadband was found to be far slower than landlines with comparable top speeds.

Overall the report found that faster broadband speeds did not equate to faster browsing for the end user. However, Sean Lockman, director of portfolio strategy in Eircom Retail, argues that much of this depends on when the user connects. “Many people will experience their peak speeds – they’ll only start to experience less than that in the evening time when everyone else is online. So I don’t think it’s disingenuous as the connection can achieve that speed most of the time for most people.”

In March 2008, the Advertising Standard Authority of Ireland issued a directive requiring the inclusion of a “busy hour” speed in broadband marketing, which would factor in contention among other things. So far none of the big players has adhered to this but Magnet Networks recently called for its rivals to do so in the interest of transparency. Providers such as Vodafone, however, say it is not a realistic option. “Advertising busy-hour speeds with any accuracy is a very considerable challenge . . . because it varies from exchange to exchange,” says Eileen Maher, head of fixed services with Vodafone Ireland. “There are huge variations nationally, and even from town to town, so it would be unfair and misleading to pick one number for all as it would be non-representative and also would change constantly.”

Vodafone specifies its landline contention ratio on its website alongside the connection’s maximum speed. However, neither Eircom nor Chorus/NTL has these details listed online.

As the authority is a self-regulatory body, it has no real powers to punish companies that fail to follow its directives. The Communications Regulator (ComReg) has no regulatory limit on contention or directives on the way it is factored in to advertised speeds but says that it advocates transparency on the matter.

Regardless, there is agreement among industry players that the advertised speeds are not reflective of the real speeds people experience, particularly at peak times.

While the broadband market in Ireland has changed since Epitiro’s survey in 2008, so have the habits of web-users, something Eircom’s Sean Lockman says is making contention more of an issue.

“We see internet video, file-sharing, online back-up and services like RTÉ’s Player becoming more popular, and they all require a more constant stream of information,” he says, citing a Cisco Systems report which recorded a 62 per cent increase in online video streaming to PCs in the past year.

“More people want to be online all the time doing a lot more stuff . . . We’re seeing a lot of our more advanced customers having three or four users on at the same time on different devices like PCs, gaming machines and smartphones.”

Simple web-browsing only requires bandwidth every time a user updates or moves to a new webpage, so it can be redistributed across the other contended connections as it is needed. Streaming services like video require a constant flow of speed, however, leaving the remaining contended connections with much less to share.

According to Lockman, Eircom has undertaken a major network overhaul to address this shift, one which will eventually remove the issue of contention for about 75 per cent of its customers.

The company has started with upgrades to the Dublin network, and plans to have all urban areas uncontended in 12-18 months.

Once complete, the upgraded lines will be available to wholesale customers as well as residential, meaning competitors like Vodafone will be able to offer uncontended packages too.

While Lockman could not give specific details, it is understood that some customers may see the benefits of this upgrade within the next two months. For others it could take some time, however, due to the extent of the aged infrastructure Eircom has to replace.

Until then, users’ broadband speeds may remain a purely aspirational figure.