NBN quietly announced yesterday that it would no longer be attempting to utilise Optus’s HFC network in its makeup. Instead, it would utilise the much touted “compromise” technology – Fibre To the Distribution Point (FTTdp) in up to 700,000 households – which effectively extends the fibre point to the house boundary and utilises copper for the final 10-50 metres. The use of Optus’s network would be restricted to the heavily re-engineered launch site in Redcliffe, QLD, and effectively abandoned outside of that area.
It’s a stunning move, and what amounts to a fairly hefty back-step in rhetoric, considering how much Fibre-To-The-Node has been used as a fall back, and actually explains a number of recent questions that arose in the corporate plan. NBN’s press release on the announcement, as they normally are, was fairly bereft of hubris, and instead made claims that this was all part of the master plan. The reality was effectively what many of us in the industry had already been saying – The Optus HFC network was not fit for service, and required staggering amounts of work to be viable as an option.
This goes some way to explaining why the number of HFC customers dropped in the recently released corporate plan – they were not going to FTTN as many had thought, but instead to FTTdp, a genuinely better option. FTTdp offers a wide range of improvements over FTTN, including (but not restricted to) a shorter copper length, a significantly higher speed cap, and a very cost effective future upgrade to fibre. It also doesn’t require a huge, ugly, exposed node and can be terminated via a book-sized box buried in the customers’ garden verge. It can also use existing copper, meaning most installs won’t go past the mailbox.
This technology was touted as a cost effective FTTN alternative by Labor in the last election campaign, namely due to its “compromise” nature, and many had figured the Coalition would be forced to use it in areas where FTTN was not viable. It’s a genuinely good decision by NBN, and drastically cheaper than reworking a tired and faulty network that had been effectively abandoned for a decade. It will provide much better outcomes for the users on it, as well as less power, engineering and maintenance costs in the long term when compared to FTTN.
Was Optus’s network really that bad?
Yes. Even before the leaked report in 2015 that detailed NBN’s internal analysis of the Optus HFC footprint and its likely high cost of overbuild, Optus had a big reputation for providing a substandard product. Nodes were oversubscribed for years, with most users getting a fraction of their promised speeds – dozens of reports and stories on broadband forum Whirlpool is a testament to this. Most of the end-user and back-end equipment was EOL (End of Life), there were high levels of noise and interference as well as transit and capacity problems. NBN noted a staggering two-thirds of the users on the network would require new end-to-end equipment.
This is all well and good, but didn’t NBN buy Optus’s network back in 2011? Isn’t this just another waste of time and money? Well, not exactly. In 2011, the government agreed to pay Optus roughly $800 million, but not for its network – for its customers. It also provided costs for Optus to progressively dismantle its old network as customers were transferred to NBN’s. In 2014, then Communications Minister Turnbull renegotiated the deal and essentially got Optus to throw the physical HFC network in as part of the deal for basically nothing. Even Optus couldn’t pretend it was an asset.
The 2014 deal effectively gave NBN the right to use portions of it as needed without paying any more money, but left the network ownership in the hands of Optus if it didn’t. As a result, Optus will still progressively migrate customers as FTTdp moves through its footprint, decommissioning it progressively. Thankfully for NBN, Telstra had been dumping quite a lot of money into its network, both before NBN arrived and afterwards, expecting that it could fetch top dollar for a sale (it did) and also later win a contract to upgrade and expand it for NBN (it did). In many cases NBN will only need to send a user on Telstra HFC a new modem, with most changes occurring at the node or further back in the network.
So what speeds are possible on FTTdp? Well, since the technology still uses copper, albiet much less of it, what’s likely is that most users at the very least are much more likely to hit speeds of 100mbits or more without technology like vectoring being necessary. How high? Currently at short distances (between 1 and 100m), real world speeds of between 600 and 800 Mb/s have been produced on G.Fast VDSL2, making this option much more appealing for users who may miss out on a potential 1 Gb/sec service over FTTP or HFC. The plus? A fibre upgrade is significantly more viable.
So while this isn’t the FTTP 180 most of us had been hoping for, it is certainly a move in the right direction. It proves that NBN does have some flexibility in providing better technical options for problems, and that its engineers are working hard to actively lower the costs of better technology that fits the new budget and rollout speed restrictions of the new regime. This backflip also breaks the glass wall of political pushback for the government – now that the FTTdp genie is out of the bag, it can be used in more places without the risk of it being wedged or embarrassed in parliament.