CVC. CR. AVC. PoI. FTTP. FTTN. FTTB. FW. Mbps. RSP. I think the National Broadband Network (NBN) has lots of people suffering from TMIS (Too Many Initialisms Syndrome – OK, you caught me, I made that one up).
My general feeling is that people don’t really care what these Initialisms stand for or mean – they simply want to have reliable and fast Internet that is commensurate with the promise from their Retail Service Provider (RSP). The ONLY reason that the general public is becoming familiar with these terms is because the NBN performance is below expectations.
That is for two reasons.
Firstly, the initial model of Fibre to the Premises (FTTP) for houses and businesses in city limits promised a world of unprecedented speed and reliability. New fibre cabling into each and every premises would replace copper that, in some cases, was over a century old. Fibre delivers a clearer signal over longer distances for telephony and is capable of data speeds many orders of magnitude greater than copper. The initial vision of the NBN was a technological masterpiece.
When a subsequent government policy change dictated that the FTTP roll out would stop and be replaced by the technically inferior and marginally cheaper Fibre to the Node (FTTN) the writing was on the wall for the first wave of disappointment. The latest quarterly ACCC NBN Wholesale Market Indicators Report shows that, for the first time, FTTN, with 1,217,212 connections has overtaken FTTP (with 1,172,836 connections). With no new FTTP roll out the gap between the two will only increase.
FTTN uses a combination of fibre optic cable to a node and then existing copper cabling for the connection from the node to the premises. Whereas the FTTP model has shiny new fibre all the way, this last section of copper is responsible for some of the speed and reliability issues and some of the initial connectivity problems that people are experiencing. Whilst the technology being utilised with these copper connections is ground-breaking and pushing copper further than anyone ever predicted it was capable of, it will never be a replacement for fibre.
Tick one for the consumer disappointment factor.
Tick two involves the Connectivity Virtual Circuit (CVC) or Contention Ratio (CR). They are both different but deliver the same result for the end user. Essentially, when you purchase a certain speed from your RSP, that RSP only purchases a percentage of that speed from the NBN. This is nothing new and has been going on since the days of dial-up and furthermore, it makes perfect sense.
This largely reflects the same model employed by a number of utilities. Your house would probably have a 19mm water pipe feeding into it. The mains for your subdivision might be 100mm. The cross-sectional area of the water mains is only 28 times greater than your house but there might be 100 or more homes on this water mains. The logic is that it is unlikely that all homes will have all taps on at the same time – and when many people are simultaneously using water (morning showers) there is a drop in water pressure. Electricity providers are the same – your house might have a main circuit breaker rated at 100A and if there are 50 homes being fed by one transformer the electricity providers don’t build capacity for 5000A as the assumption is that not every home is using its full capacity at the same time.
Rather than delve into the nitty gritty of the new charging model for the CVC and the discount structure put in place, the contention ratio is the easiest way to understand our bandwidth. The latest ACCC report shows that the RSPs in Australia have sold 94,240Gbps of bandwidth to their customers. They have purchased a combined 3,364Gbps of bandwidth from the NBN (average weighted speed is 31Mbps). That is a contention ratio of 28:1. Across the world you will see contention ratios of anywhere from 20:1 to 50:1 with a lower number being better so Australia is at a reasonable level.
As with water and electricity, if everyone used their maximum capacity at the same time, we would all only receive 3.6 per cent of our promised speed (some RSPs have better ratios but I am averaging this out across Australia). In the past, when our Internet usage involved browsing the Web or answering e-mails, the very nature of Internet usage was in small bursts so a 28:1 contention ratio was more than adequate. In fact, it was outstanding.
Then came Netflix. More accurately streaming but Netflix is the major player in this space. Where the second phase of dissatisfaction arises is when Australia sits down after their evening meal around the TV. They no longer watch free to air – they stream. And streaming does not use short bursts of data – streaming uses a constant stream of information and what was once entirely acceptable with a 28:1 contention ratio is now proving to be somewhat inadequate.
The solution is for RSPs to purchase more bandwidth (called CVC) from the NBN. This comes at a cost though which is where the new pricing model from the NBN may help. To make it tougher, RSPs don’t publish their contention ratios so it is difficult for consumers to make a decision based on price and performance.
Watch this space as I think we will see more changes for Australians to have their NBN dreams fulfilled. Or we could just stop watching Netflix – but I don’t think that is going to happen anytime soon.
Mathew Dickerson