» ISP News » 

Future 1Gbps FTTC Broadband G.Fast Tech Passes First Stage of Approval

Tuesday, July 16th, 2013 (9:28 am) - Score 4,010
fttc rural bt street cabinet

The next generation of FTTC (VDSL2) technology, which is better known as G.Fast and could one day deliver peak internet speeds of up to 1000Mbps (Megabits), has passed its first stage of approval on the road to becoming an official ITU standard in early 2014.

The current generation of 80Mbps capable Fibre-to-the-Cabinet (FTTC) technology, which is being rolled out by BT, is expected to largely replace or compliment standard ADSL / ADSL2+ connections as the dominant form of UK broadband connectivity over the next few years.

FTTC works by running a fibre optic cable from the nearest telephone exchange, which is then taken to your local street cabinet. After that homes and businesses are connected by using VDSL2 technology over the “last mile” run of existing copper cable.

G.Fast is often seen as the next generation of FTTC, which could push speeds of up to 1Gbps into homes. But this would only help those living either very close to their street cabinet (i.e. up to a distance of around 250 meters) or to a fibre optic distribution point (FTTdp). The recent Alcatel-Lucent trials give a rough indication of what it could do (here).

But G.Fast first has to become an official standard. The ITU today reports that G.Fast has passed through the first stage of this process after gaining approval for a recommendation (ITU-T G.9700) that specifies methods to “minimize the risk of G.fast equipment interfering with broadcast services such as FM radio“.

Hamadoun Touré, Secretary-General of the ITU, said:

Since the early days of the World Wide Web, people around the world have accessed the vast resource that has become the Internet via ITU standards. I applaud our membership for continuing to show great leadership in the development of these specifications that bring broadband into our homes at ever increasing speeds and at ever greater efficiencies.”

Tom Starr, Chairman of ITU-T Study Group 15 (Working Party 1), added:

G.fast is an important standard for service providers globally. [ISPs] will be able to deliver fibre-like performance more quickly and more affordably than with any other approach.”

Interestingly the ITU states that G.Fast will also enable self-installation by consumers without a technician’s assistance, which is already something that BTOpenreach will shortly begin to offer alongside standard FTTC broadband connections; albeit with some expectation of performance loss due to the use of special filters as opposed to an engineer install.

It’s hoped that the new standard will gain full ITU approval during early 2014 and it will then need to be tested and trialled by all of the usual suspects (e.g. BT are already experimenting with it in their labs). However G.Fast would be both expensive and complicated to roll-out, which means that we’re unlikely to see related products for a fair few years yet. Vectoring will come first to help reduce crosstalk (interference) and improve existing speeds.

The following is a recent ITU video in which the new standard is discussed.

Add to Diigo
Tags: , , ,
Mark Jackson
By Mark Jackson
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he is also the founder of ISPreview since 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on Twitter, , Facebook and Linkedin.
Leave a Comment
41 Responses
  1. Slow Somerset

    Getting the Current Generation would be a start here In Somerset.

  2. JNeuhoff

    This is just attempt to delay the inevitable end of copper: Full FTTP which is already capable to do up to 1gbps

    • FibreFred

      Or to put it another way (the correct way) this is Alcatel-Lucent’s proposal to extend the extensive copper plant investment across the globe just like ADSL and its variants and VDSL and its variants before it.

      Makes perfect business sense for some areas, obviously you would have jumped straight from 14.4 dial-up to FTTP 😉

      … and quickly gone out of business

  3. Clive

    FTTC at 1Gbps in early 2014!!!! I will believe that new sound bite when i see it. Still waiting on so called vectoring, increased profiles and other FTTC speed boosts.

    • JNeuhoff

      This G.FAST won’t be around for another 2 or 3 years. By then it will be pretty obvious that genuine fibre is superior to all these little copper fixes.

    • MikeW

      Everyone agrees that true fibre to the premises is technically superior to all the copper variants – FTTC, FTTS, FTTdp and HFC included. No more “up to” that depends on distance, lower maintenance costs, and higher speeds – provided ISPs choose to sell the higher packages.

      The one way that fibre isn’t superior to copper is when the copper plant is already in the ground – the financial penalty to have to pull/blow fibre to every doorstep cannot be underestimated.

    • FibreFred

      Everyone knows fibre is superior, its known now.

      As MikeW says if this is brought to the pole it could bring most fantastic speeds, I’m about 15-20metres of cable to my pole, I suspect most people are around the same distance.

    • Clive

      Agree entirely JNeuhoff 🙂

      @Fibrefred I also agree that 20(ish) Metres of cable from house to pole is probably a good average, which makes things even more weird to have a technology that does not bother to replace that final 20 Metres of copper but replaces sometimes several miles of copper before it. Seems a bit silly?

    • FibreFred

      Not really, if you can get comparable speeds to fibre it works out a cheaper install, a set of fibres to the pole, hook up the DP and hook up the copper pairs on the pole, no need to visit the house to run your fibre in.

    • Clive

      Erm i do not think so. Current FTTC and other fibre/copper mixed products over your telephone cable still require an engineer to visit your home (though i admit that may change soon if/when BT allow users to install their own filters).

      Even if that happens they still need to visit a cabinet. In addition Filters are still needed. Which i assume will still be supplied by BT.

      Also if you are terminating the fibre cabinet at the pole and then running copper from that (IE the drop wire) to the persons home does that not require some kind of termination equipment at the pole in which to send the signal from the fibre cable into the copper cable??? What is the cost of that equipment???

      Do not see how you think that is is cheaper over just continuing the fibre run to the persons house. Fibre cabling costs as little as £1 a metre nowadays. So for that extra 20M you are looking at £20….. Somehow i think that is cheaper than additional equipment on a pole to connect the fibre into the copper and supplying people filters.

      Its another product that makes no sense and will probably never be used.

    • Mike

      The only people that have ever stated FTTC would cost more than FTTP here in the UK is BT thereself.

    • FibreFred

      “no need to visit the house to run your fibre in.”

      I said fibre

      What would need to be fit in the home, maybe a new filter, maybe a new CPE? FTTC is moving to self installs, no reason why G.Fast couldn’t

    • Clive

      ““no need to visit the house to run your fibre in.”

      I said fibre

      What would need to be fit in the home, maybe a new filter, maybe a new CPE? FTTC is moving to self installs, no reason why G.Fast couldn’t”

      Which makes no difference to labour costs as a bloke still has to go to a cabinet to enable a persons FTTC connection, a cabinet which in most cases will not be far from a persons home anyway.

    • FibreFred

      So engineer going to the pole to fit fibre and then finishing up because the customer can fit the CPE themselves


      Engineer going to the pole to fit fibre, then running fibre from the pole to the house, drilling through the property, fitting the ONT and BBU and voice link to the customers internal voice cabling

      There’s no labour cost different between the two?

      Might want to have a think again about that one “Clive”

    • MikeW

      BT’s experience of installing FTTP has lead a two-stage process. One booking does the external work, to bring everything close to the house, while the second booking does the final drop into the house.

      IIRC the second booking takes an average of 7 hours, while the first one is longer (and with multiple people).

      Against a 1 hour appointment to install an FTTC filter and modem.

      I can’t recall where I saw it, but an estimate for the cost of installing FTTP were that 80% comes from the final 200 metres into every individual house. Partly because of the sheer number, and partly because of all the time needed to organise appointments.

      Heck, I’d never even heard of “FTT-Wall” before, but it is a solution where the copper still goes from an NTE on an external wall to the inside of the house – to save the need for the telco engineers to enter the property. Daft as the idea might seem, if it is being talked about, then there are obviously time & cost advantages to not needing to make appointments.

    • Clive

      As already pointed out to “YOU” BT already give away 30 etres of cable with FTTC and can spend several hours installing it, costs would be no different. Unless of course BT have some sort of dubious method that charges more for x amount of hours work installing one product over another which they are fond of doing.

  4. MikeW

    VDSL2 was standardised in 2006, and took 3 years to become “acceptable” by the UK’s regulatory bodies as well as becoming a strategic product direction by BT. Another year for the first “widespread” public trials, and another 2 years to become a technology available to a majority of the public.

    Vectoring was standardised in 2010, and is commonly available in tested hardware about 3 years later. Same year for initial public trials, and probably a couple of years away from being a widespread solution in the UK.

    If G.Fast follows the same timescales, we should expect private trials in 2015-16 (and lots of hype), public trials in 2017, pith it perhaps becoming available widespread in 2019-2020.

    That is if BT see it as a strategic product for their portfolio at the time. FTTP and FTTPoD might have such an amazing take-up that BT could decide to skip G.Fast.

    Fibre-lovers will enjoy one thing about G.Fast and FTTdp – it *will* be the last hurrah for CTTH.

  5. Ignitionnet

    This is a while away in any event. First there’s vectoring, then there’ll be pair bonding to get more out of the existing FTTN architecture.

    Vectoring, pair bonding and phantom mode technology deliver 300Mb at 400m via 2 copper pairs.

    • JNeuhoff

      Dream on. Bonding technology is a nightmare to use, we did it for many years. Also, bonding multiple copper lines means multiple redundant voice services, hence hopelessly overpriced. Also, additional copper lines are not always available. Money should be spent on proper FTTP solutions, not half-baked copper workarounds.

    • Clive

      Agree bonding on FTTC will never likely be available, the only ISP i know of that even offered ADSL bonding was BE and i believe a couple of others tried it but it never really caught on.

      Vectoring also looks like another day dream, the first mention of that was way back end of 2010, still nothing except very limited tests.

      We are going to be stuck with 80Mb FTTC for years more yet.

    • MikeW

      When Bonding is mentioned for FTTC, people forget that the technology has jumped from the old ADSL bonding solutions. Older solution may well have needed multiple telephone services (each with DSL) to be purchased, with the bonding bolted on.

      (And there seem to be plenty of business ISPs that still offer such a solution)

      The newer services can happen direct from the DSLAM, only requiring the extra copper pairs. Bonding can be implemented within the scope of the purchase of a single phone service with FTTC DSL but with multiple copper pairs – if BT ever chose to offer it.

      Of course, we still have no idea whether BT ever to intend to do this.

    • MikeW

      Vectoring, meanwhile, is following the usual timescales of standardisation. Once approved in the ITU (which was 2010 for vectoring), it takes a few years for the hardware and software to develop sufficiently into products worthy of mass deployment, and the uptime requirements of operators.

      The hardware companies spent 2010-2011 performing trials of prototypes. 2012-13 has seen real hardware (and better speeds), alongside multiple interoperability plugfests – all needed before you can get to live trials.

      Don’t confuse the time needed for development work to mean the technology is a dream. Vectoring is going neither too fast nor too slow through the process. About the same speed as VDSL2 did.

    • Clive

      An example of where these “Better Speeds” are please

  6. New_Londoner

    The fibre fetishists seem to miss the point. It is not whether FTTH is better, it is simply the cost and time to deliver it to every premise versus the (much lower) cost to upgrade cards in cabinets etc.

    If G.Fast provides a route to deliver a few hundred mbps to most properties this decade, why would you want to spend say £25bn putting in FTTH? Remember a lot of the FTTH cost is manpower, so is unlikely to reduce, as opposed to the cost of THE ELECTROINIC

    • New_Londoner

      ^^^^ CONTD – hit return in error

      Remember a lot of the FTTH cost is manpower, so is unlikely to reduce, as oppossed to the cost of the electronics for G.Fast and its ilk, which will almost certainly come down as scale picks up.

      You could easily envisage this pushing out timescales for mass deployment of FTTH so its always a decade away, unless there is a surge in orders for fibre on demand. But this seems unlikely if bandwidth availability stays ahead of mainstream uses, as is the case with FTTC and HD TV at the moment.

      Of course there will be exceptions, but its difficult to see these becoming mainstream. Its even more difficult to see a rush of investors willing to put in the money for a national FTTH programme, even more so for the government given the strains on the public purse.

      Is this the last hurrah for copper? Well it wasn’t that long ago when getting Ethernet over twisted pair within a building was seen as difficult, so I wouldn’t be too surprised to hear about G.Fast2 or G.Faster (or even G.Superfast) as the ink is drying on the new standard.

    • MikeW

      I think the DP is the last point where you can make economies of scale – one node per 10, 20 or 30 subscribers. At the DP, the individual copper lines spread out to the individual properties, so to put an active electronic node closer to the home than the DP becomes a 1:1 thing.

      It isn’t impossible though. The Alcatel-Lucent article on G.Fast mentions a concept of “FTT-Wall” which *is* a 1:1 device (copper from the outside wall of the property to inside) – it doesn’t offer economies of scale, but it does get around the need for making appointments with householders.

      So I tend to agree – copper may last longer than we think. People seem to be happy with a final copper segment of gigabit ethernet, because it (nearly) matches the capacity of the fibre. For example – no-one complains about Hyperoptic, but theirs is an FTTB solution that does exactly this. FTTW with G.fast for the final few metres would perform this.

  7. Clive

    “The fibre fetishists seem to miss the point. It is not whether FTTH is better, it is simply the cost and time to deliver it to every premise versus the (much lower) cost to upgrade cards in cabinets etc.”

    The only problem with that statement for future Fibre/copper mix solutions is it is irrelevant.

    Once you start moving to tech where the fibre runs all the way to the pole rather than a cabinet and then the only bit of copper is the drop wire from a pole to a persons house there is hardly any “cost” saving there.

    In fact depending on how it is deployed and what additional equipment is involved (another box on pole of some description to feed the fibre to the copper, new filters in users home etc etc) it could as ive pointed out elsewhere cost more than just running fibre all the way to the house anyway.

    Unless of course it will be another scam product where BT think they can charge an end user thousands to have it installed like they hope they will be able to do with FTTPoD and it never catch on.

    • JNeuhoff

      “Unless of course it will be another scam product where BT think they can charge an end user thousands to have it installed like they hope they will be able to do with FTTPoD and it never catch on.”

      BT’s so-called FTTPoD isn’t even available for most parts of the UK, it is still in the “trial” stage. You won’t see this product for a long time to come, it was more of a BT publicity stunt!

    • New_Londoner

      Not sure about your comment that there is “hardly any cost in replacing [27 million?] drop wires”! The amount of labour involved is high, it’s the labour cost not the parts that needs consideration, something that technologists and enthusiasts usually overlook.

      If there was a genuine business case for FTTH then surely investors would be clamouring with their £25bn cheques? Unfortunately for you this is not the case, so where is your evidence that this is a viable investment, in the UK or elsewhere?

    • Mike

      If you have a solution where fibre runs as far as the pole and then copper runs from the pole into a persons home, then surely the labour cost of that is just the same as each time a person takes the product some bloke has to climb up a pole and connect it?

      The labour cost would be no different if you had fibre running direct into the home they would have to connect that to equipment in your home (probably a 30 minute to 1 hour job) which is likely to be the same time it currently takes for an engineer to connect you up on FTTC or a system where they connect a copper wire to a fibre wire at the pole.

      Where is this extra labour you talk about?

    • Mike

      As to cost of replacing drop wire. BT already give away cable which would equate to the cost of that with FTTC installs. They already have a ‘home wiring solution’ which involves a BT engineer running UPTO 30 Metres of cable around your home.

      I agree with Fibrefred about the average drop wire being around 20 Metres and i agree with Clive about the cost per metre of Fibre cabling (IE around £1 per metre).

      A 30 Metre Network cable which is basically what BT’s ‘home wiring solution’ is, can cost anything from £10 to £40 a face plate is also included at a cost of about another £2-5 and they have no issues giving that kit away FOR FREE.

      So why would they have issues replacing 20 Metres of copper drop wire with 20 Metres of fibre? Certainly is not cost.

    • Clive

      Well explained about labour costs.

      Oh and very, very, very good point about the BT Home Wiring Solution, i never thought about that. They already include an additional spend on 30 Metres of cable already and another faceplate i did not think about.

      @JNeuhoff Well i was trying to be partly fair, but yeah FTTPoD, along with vectoring, additional FTTC profiles and so many more “speed increase” promises is indeed another thing that can not yet be had.

    • MikeW

      “If you have a solution where fibre runs as far as the pole and then copper runs from the pole into a persons home, then surely the labour cost of that is just the same as each time a person takes the product some bloke has to climb up a pole and connect it?”

      If the box were installed once, and every line physically connected, then you just have the single installation to figure into the labour costs.

      As I mentioned before, it appears that 80% of the cost comes from those final few metres – the ones where you’d have to visit every individual property, making appointments each time.

      If you want to see a breakdown of costs for different FTTx architectures, take a look at figure 5 in this Alcatel Techzine article on G.fast. I guess the dark-blue (civils) and orange (home installation) bars are the ones where the heavy labour costs appear.

      “Well i was trying to be partly fair, but yeah FTTPoD, along with vectoring, additional FTTC profiles and so many more “speed increase” promises is indeed another thing that can not yet be had.”

      All technology starts off this way, and takes time to develop suitably from individual lab prototypes into robust, reliable hardware for millions. And more time to deploy to millions. Is it truly being fair to poo-poo things that are still heading down the development path? The *expected* development path? Still going through the *expected* deployment path?

    • MikeW

      “So why would they have issues replacing 20 Metres of copper drop wire with 20 Metres of fibre? Certainly is not cost.”

      It isn’t the cost of the cable. It is the cost of the labour, and the ancilliary cost of having to make (and honour) appointments for individual homes, and travel between those appointments. They’d save money if they did homes in batches, and forced you to be present at a time that helped them.

      I think the average drop-wire length is nearer 60 metres, if I remember correctly from some BT presentations. While it triples the cost of the cable in your calculations, it is still only a small component compared to the labour cost.

    • Clive

      “I think the average drop-wire length is nearer 60 metres, if I remember correctly from some BT presentations.”

      You do not remember correctly and like most of what you have to say that is clearly made up.
      Maximum length for a dropwire10 is 68 metres. DW10 is 2 pair.
      Maximum length for DW14 is 55 metres. DW14 is 4 pair.

      SO the average of 20-30 Metres as already discussed is far more likely to be accurate.

      If you have nothing to contribute rather than blindly defending an organisation with out researching anything its probably best you do not bother.

  8. neilfairbrother

    FTTdp – or fibre to the telegraph pole. Where does the power come from?

  9. zemadeiran

    How about eliminating poles from the equation and just run the fibre along the gutter in a terrace?

    That would sort terraced deployment but I can see an issue in regards to private building work etc.

  10. No consideration to the EMC implication is given. The mask is
    up to 424 MHz with only Band ii proposed for notching; the risk
    of radio interference to both broadcast (e.g. DAB) and other
    legitimate (authorised) radio users is significant. The uptake
    of 80M FTC has already resulted in an increase in radiated

    The radio spectrum is a finite resource. Just recently a museum
    in Norway was forced to turn off its LED lighting due to
    interference with the civil aviation band. G.fast makes no
    mention of notching this band – or any other, except for VHF FM
    which, depending on the Culture Secretary on December 16 2013
    may be scrapped any way. . .

    The inane lust for data bandwidth and lack of market
    surveillance is destroying the radio spectrum. The existing
    copper network is not suitable for transporting radio frequency
    energy which inevitably ‘leaks.’

    The EMC regime in Europe is broken but was put in place for a
    good reason; in the 1980’s radiocommunications were decimated by
    the explosion of personal computers with little or no screening,
    anyone old enough to recall driving down their local high street
    in that era will recall the effect on their car radios. . . The
    “need” for more data bandwidth is taking us there again, only
    this time the spectrum regulators appear to have had their hands
    tied to prevent their intervention on the grounds of DG
    Enterprise & Industry using their enormous whipping stick
    called “there shall be no barrier to free trade.”
    The radio spectrum is a natural resource and is equally
    sensitive to pollution as any oil spill can have on marine life.
    The proposal to implement G.fast onto a dense copper network,
    concentrated in towns and cities, with long, efficient radiating
    elements (telephone cables) will have a drastic  effect, no
    matter what G.fast lobby would tell you otherwise. The same
    thing has happened with powerline networking which is already
    causing untold harm and which is facing a 10,000 x power
    increase with the implementation of the recent EN50561 which
    demonstrates how broken the system really is: a Standard which
    breaches the Essential Requirements and thus is unlawful. . .
    In 911 the telecom infrastructure broke down with the loss of
    the oversubscribed cell tower on the collapsing building. A
    major event such as a modest solar coronal mass ejection toward
    earth is all it would take to knock out the complex telecoms
    infrastructure. When events happen like these, the go-to service
    for immediate support is the amateur radio community which is
    now haemorrhaging users due to the already untenable
    interference problems being caused by the existing telecoms data
    mechanisms. Even GCHQ stated that it was affecting them – a
    statement which was withdrawn and hushed by the UK government under
    a special order
    Please do not dismiss the importance of the radio spectrum or
    its users nor underplay the impact of the current wave of
    interference being driven by greed and profit. Increasing the
    speed and bandwidth of data transmission mechanisms relying on
    copper transportation is unethical and harmful to the
    environment – the EMC environment. Just because an electrical
    system claims to be balanced and interference-free on paper,
    does not mean that it is in reality. If this were the case then
    VDSL would not be causing the problems that it is.

IMPORTANT: Javascript must be enabled to post (most browsers do this automatically). On mobile devices you may need to load the page in 'Desktop' mode to comment.

Comments RSS Feed

* Your comment might NOT appear immediately (the site cache re-syncs periodically) *
* Comments that break our rules, spam, troll or post via fake IP/proxy servers may be blocked *
Cheapest Superfast ISPs
  • Origin Broadband £23.61 (*31.58)
    Up to 38Mbps, Unlimited
    Gift: None
  • Plusnet £24.99 (*33.98)
    Up to 38Mbps, Unlimited
    Gift: None
  • Vodafone £25.00
    Up to 38Mbps, Unlimited
    Gift: None
  • TalkTalk £26.50 (*33.50)
    Up to 38Mbps, Unlimited
    Gift: None
  • Pop Telecom £26.99
    Up to 38Mbps, Unlimited
    Gift: None
Prices inc. Line Rental | View All
*Javascript must be ON to vote*
The Top 20 Category Tags
  1. BT (1985)
  2. Broadband Delivery UK (1352)
  3. FTTP (1308)
  4. FTTC (1262)
  5. Openreach (996)
  6. Politics (990)
  7. Business (888)
  8. Statistics (795)
  9. Fibre Optic (765)
  10. Mobile Broadband (721)
  11. Wireless Internet (660)
  12. Ofcom Regulation (651)
  13. Virgin Media (605)
  14. 4G (605)
  15. FTTH (548)
  16. Sky Broadband (468)
  17. TalkTalk (443)
  18. EE (389)
  19. Security (322)
  20. 3G (280)
New Forum Topics
Helpful ISP Guides and Tips

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms  ,  Privacy and Cookie Policy  ,  Links  ,  Website Rules