Home
 » ISP News » 
Sponsored

Broadband Speeds of 10Mbps Needed for an Effective Quality of Service

Friday, October 31st, 2014 (11:09 am) - Score 3,090
broadband internet speed speedometer

The United Kingdom’s telecoms regulator, Ofcom, has suggested that there is an “increasingly compelling argument” for fixed line broadband download speeds of 10Mbps (Megabits per second) to be considered a requirement for an “effective quality of service,” especially where there is simultaneous use of the connection by different services or users within the home.

At present the Government’s official policy as part of their Broadband Delivery UK (BDUK) programme is to make “super-fast broadband” speeds of greater than 24Mbps available to 95% of the population by 2017, with download speeds of at least 2Mbps also being available to everybody as part of a non-legally binding Universal Service Commitment (USC).

But Ofcom’s new Citizens and Communications Services report, which was published today and looks at the availability, accessibility and affordability of communications services across the UK (note: nearly all of the statistics it uses have already been published), claims to have identified three key “coverage challenges” for the future.

Fixed Broadband Availability (Ofcoms Top 3 Coverage Challenges)

1) A small but significant number of homes are still unable to access a service of at least 2Mbit/s, and there is more work needed to deliver wider availability of good quality broadband, especially in rural areas.

2) There is an increasingly compelling argument that speeds of 10Mbit/s are needed to achieve an effective quality of service, especially where there is simultaneous use of the connection by different services or users within the home.

3) Superfast broadband coverage has increased rapidly, but remains short of universal coverage, and there are challenges around rural coverage and also some urban ‘not-spots.’

Admittedly Ofcom, which are somewhat stating the obvious above, then goes on to remind us that 78% of UK premises can already order a Next Generation Access (NGA) broadband connection and 8% of broadband connections operate at speeds of less than 2Mbps; although over half of those already have the option of switching to a supposedly “superfast” service. Take note much of this info. is from their old June 2013 data and the regulator will be releasing new information very soon (thank god).

download_sync_broadband_speed_distribution_ofcom_2013

Interestingly Ofcom, which since as far back as 2012 has been suggesting that the 2Mbps USC should perhaps be lifted to 8-10Mbps (here and here), now indicates that this could instead form the basis for a third level of broadband performance to gauge the quality of consumer services.

Precisely what this would mean in practice is not yet clear, although we suspect that it’s more of a change for Ofcom’s own reporting than one that could impact Government policy. One of the reasons we say that is because the Government has retained their existing USC level of 2Mbps, despite Ofcom’s earlier suggestions that they adopt something faster.

On top of that the BDUK definition of “superfast” (24Mbps+) still differs from Ofcom’s definition, which much like the EU pegs the figure at 30Mbps+ (note: some newer BDUK contracts do use the 30Mbps figure).

Ofcom’s Report Extract

Moreover, the broadband speed required by a typical household is likely to increase over time, as individual applications require more bandwidth, and the number of simultaneous applications increases. We may begin to view around 10Mbit/s as the effective quality of service broadband consumers expect, with 2Mbit/s more of an ‘essential’ level. Last year we estimated that around half of all broadband connections in the UK were at least 10Mbit/s.”

Otherwise Ofcom’s new report merely appears to re-hash old data that can be found in their various other studies, such as the Fixed Broadband Speeds Report, Infrastructure Report and Communications Market Report (CMR). Ofcom’s next major Infrastructure Report is due out before the end of the year and that will be a more productive read.

As to whether 10Mbps does give consumers an “effective quality of service”, well that will invariably remain a matter of debate. Granted you can do most things with 10Mbps, but with HD video streaming becoming common place and 4K likely to follow, not to mention those 50GB game downloads, then we can already see it being put under strain.

Ultimately everybody has their own unique requirements and expectations, yet if we’re going to start saying that people need 10Mbps for an effective connection quality then perhaps we should just make that the national USC and get on with the job of making it possible. Alternatively we could consider 24Mbps+ to be the USC, especially as that’s going to become the eventual aim of all this deployment work.

Share with Twitter
Share with Linkedin
Share with Facebook
Share with Reddit
Share with Pinterest
Mark Jackson
By Mark Jackson
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on Twitter, , Facebook and Linkedin.
Leave a Comment
34 Responses
  1. Avatar DTMark

    Why on earth would we need 240v electricity?

    Pretty sure many could get by with less than that by simply not using too many devices at the same time.

    Seems insane that we looked into the future for one moment way back when.

    Having had access to a varying range of speeds over the years, the absolute bare minimum for the internet to be even faintly useful to me is about 10Meg down 3Meg up.

    For a family, I’d have thought you’d be looking at more like 50Meg down 10Meg up. These are not fast speeds and are perfectly achievable with a competent network and suitable technologies.

    Now bearing in mind that infra is not about the now, but the future – takes time to build, factor it forward to say 2020 and considering video streaming and gaming, I’d have thought we should be looking to 100/20 which again is not that fast and perfectly achievable if we built a broadband network.

  2. Avatar Jamie

    Personally I would like to see upload speeds matching download speeds preferably 100/100. Cloud based services such as backup need a good upload to be a viable solution.

  3. Avatar gerarda

    When the USC was first mooted it was about 50% of the average download speed, so Ofocm are going in line with that. I think a USC of 50% of average download speed would be a better target than some arbritary number that is likely to be soon outdated.

  4. Contended services are useless at these speeds.

  5. Avatar gerarda

    Ofcoms reminder that over half of those of less than 2mbps already have the option of switching to a supposedly “superfast” service was based on their belief that FTTC speeds did not deteriorate with distance from the cabinet.

    A recent pronouncement suggested that they had begun to understand that this is not the case, but appears the lesson has been forgotten

    • Avatar Patrick Cosgrove

      Ofcom also needs to understand that whatever speeds people in rural areas were getting a couple of years ago are now considerably less. Mark says that Ofcome is stating the obvious, and maybe they are, but it needs to be stated, repeated ad then shoouted at this stone-deaf government that continues to trumpet highly debatable successes about rural roll-out.

    • 2 Mbps – was sufficient (forgive the notion of a limit) to support a quality home working experience for 1 person, but not enough for an IPTV service, or a house with kids, whose half term homework need the uploads speeds associated with a design studio. 1 piece of History homework half term this week is a 70 MByte file – a 14 year old, where is more than I have ever done, sound and music overlays, video and image inserts – so we had to zip it.
      I want to email in advance to apologise for files more than 1 MByte, it’s the equivalent of my dad shouting at my sister 40 years ago to get off the phone. I apologised to my son this week for nearly stopping him from using an app he was using to do some really detailed route planning for a cyling trip. He knows the cadences, altitudes of most of alpine climbs by heart and target times. I hope there is a GCSE in this topic.
      So 2Mbps you can homework but not sell your home. 10Mbps is the new 2Mbps with the same limitations if you have kids or just using your kit to its potential.

  6. Edge of network it would be ok, but their own research from Mediatique on Future TV concludes that for on-demand TV more than FTTC is needed. I would recommend the read – 70 + slides

    Furthermore if your supporting cost recovery on copper using replacement costs, then why not plan fibre which has a lower long term cost over 15-25 years. You could at least fire the gun.

    If they think this and the matter was raised in the FLAM consultations by several LA’s why did they ignore it?

    • Avatar Steve Jones

      Apart from your dubious assumptions that copper is replaced every 15-25 years, and that finance could be found, there’s a very good reason why copper can’t just be ripped out, and that’s because the regulatory and legal environment won’t allow it (apart from the odd very small scale trial in a rural area).

      Also, for a long period in any one area, there will be parallel networks running in order to maintain service. It’s taken several years to do the swap-over in the Island of Jersey (which is not complete), and both networks had to be kept going at the same time. Jersey is also a compact location, with a relatively simple legal and economic environment (and not constrained by EU rules).

      There are dozens of academic articles on the economic issues of migration, and you seem to ignore all of them.

    • Avatar GNewton

      Steve, your reasoning is quite flawed here.

      The initial costs for installing a copper line are recovered multiple times just by the line rental alone over a period of e.g. 25 years.

      As regards Jersey being a compact location, with a relatively simple legal and economic environment: The same could be said of any local district council in the UK. what you need to compare are population densities, these are usually quite high in the UK, even in smaller towns and villages.

      You whole reasoning here sounds more like a typical ‘Can’t Do’ attitude here, nothing else. There is nothing on the part of BT to do proper telecom services if it wanted to, EU regulation would not have prevented BT from doing telecom (which BT doesn’t do anywhere in many smaller towns (copper DSL doesn’t count here!).

    • Avatar Steve Jones

      @GNewton
      Your principles do not accord with the way Ofcom calculate their regulated costs. It allows for only a modest rate of return on the capital employed. The rest goes on actual costs. That means paying for things like IT systems, rates, rental on buildings, maintenance of infrastructure (poles, ducts, cabinets etc.), fuel, vehicles, power, maintenance contracts, overheads, consumable and all the manpower costs associated with repair. Of course, there is an allowance for replacement costs (and the cycle depends on the asset type), but that covers a lot more than just copper, and doesn’t include digging up everybody’s front garden every few years. It certainly does not cover the massive extra costs of running parallel networks for the five years or so that would be required in each area (taking Jersey as the example).

      As for your comparison of the legal powers of councils with the governance of Jersey, that’s a joke, right? Councils have absolutely zero control over the regulation of telecommunications save the limited powers they have over planning and street works. They cannot override national or EU laws. They cannot approve the removal of copper networks, any attempt at which would cause immediate referral to Ofcom, national and even European courts.

      The most likely scenario is that there will be a gradual, and incremental extension of fibre deeper into the network augmented by fibre to new developments, some (state supported) rural areas and, just possibly, some cherry-picked areas.

      However, whilst there is a regulatory requirement to maintain copper in existing networks, there is a huge barrier. Both because it’s cheap (sunk costs) so presents a competition issue, and because it means that the costs of both fibre and copper have to be sustained.

      To repeat again. The regulatory and legal environment does not allow the copper network to be withdrawn.

    • @Steve – We will have or we are ceratinly paying for fibre to be delivered deep into the D side of the exchange, and all fibre is proving more cost effective where power costs are excessive.

      The cost settlement is based on replacement costs and of the c£86 a year £17 is for the drop wire which the FLAM documents state as a life span of 10 years and is costed accordingly.

      Parralel running for sure but do not suggest more could not be done.

      BT is collecting circa £2.2 bn cash through this cost recovery mechanism, but Openreach is reducing its capital expenditure while having the highest level of labour capitalisation of any telco in Europe.

    • @steve The product definiton for PSTN access includes FTTP supported with an ATA, so it can be pushed now and developed further.
      FLAM can be reviewed mid-term during 2014-2017 if there is the appetite.

  7. Avatar GNewton

    Steve, as I said yours sound like a lame excuse in the interest of a BT shareholder.

    Users should campaign for the removal of potential legal obstacles if these truly prevent a telecom company like BT from doing a proper job. Shouldn’t be too hard.

    A good starting point is the replacement of Ofcom with a genuine telecom regulator who would actually do its job. And a complete splitup of BT, where the network provider becomes a genuinely independent company, not owned by the BT Group. Add to this a legislation to prevent the building of parallel networks paid by taxpayer’s money. Fibre has to replace copper, not run in parallel to it. Almost everything which could hav done the wrong way, has been done the wrong in the UK. And that includes the Cannot-do culture here.

    • Avatar Steve Jones

      “Users should campaign for the removal of potential legal obstacles if these truly prevent a telecom company like BT from doing a proper job. Shouldn’t be too hard.”

      Really, when there’s a single voice on one side and several companies on the opposite? LLU operators have significant investment in copper, and I can’t imagine them accepting the higher line rental that will be required to finance the extra investment in the project and the parallel running that would be required. The one-off capital costs would not be less than about £25bn (or about £1k per premises), and that would have to be amortised over a period. Add in financing costs, then you are talking of £4-5 per month for 20-25 years. At that cost, expect a lot of defection cable. Public money isn’t an option either where the cable network is to be overbuilt, as that will simply end up in court.

      Those who think that all this can, somehow, be covered using existing line rental costs are living in fantasy land.

      As for breaking up BT, of course that’s easier now (at least the Openreach bit), but it doesn’t get rid of all the potential conflicts of interests. For instance, FTTC cabinets are operated by OR (as they have the workforce in the field). So that’s potential competition with customers. Where do the DSLAMs go? Would they become BTR assets, effectively turning BTR into an LLU operator? Where would that lead resellers of those products?

      There are also little issues like how the pension fund liability would be distributed. The great majority of BT pensioners will have worked in the parts of the business that are now either OR or BTW, but some would have to be allocated to BTR. I can’t imagine the government being very happy to pick up the tab (as they are liable) if any of the successor companies went broke.

      New Zealand has done something like this, but it’s actually cost the state a lot of money, and it’s a vastly simpler industry structure. (Although it’s not the enormous mess of the Australian NBN).

      As for Ofcom, not being a proper regulator, what is it that you particularly object to in what they do? Their remit was to encourage competition as deep into the network as possible, reduce BT market share and minimise retail costs through competition. Unlike some countries, their has been no serious questioning from the EU over the extent of telecoms regulations in the UK. Indeed, it’s a very intrusive regulatory regime by any standard.

    • Avatar GNewton

      Mentioning the Australian example (which later went terribly wrong):

      The original NBN project to just build a single fibre-optic network for almost the whole of the country, not multiple parallel networks, was the right one. Costing-wise it was treated as a longterm investment, where the initial public funding would have been eventually recouped through revenues coming off this new network, and by getting other commercial shareholders onboard. And it was designed to REPLACE older copper-networks and cable networks. Along with a proper regulation (ACCC) it would have worked, and market competition would come from several service providers (e.g. voice services, video streaming, TV, broadband, public health service, etc). Alas, after the change of government about a year ago, it all went downhill.

      The newly elected government got the notion to change it all, to copy the poor UK example, with mad ideas like multpile technology mix, parallel networks, 25 Mbps for all via VDSL (which was never even possible), virtual abandonment of building future-proof fibre networks, now burdening the taxpayers via a public gap-funding models which would mainly benefit one monopoly company (Telstra). In short, Australia is now doing all the mistakes done in the UK, only the Murdochs will be happy now 🙂

    • Avatar FibreFred

      Many countries are rolling out VDSL, maybe its not a mistake, maybe the mistake is your own personal viewpoint?

      Just a thought

  8. Avatar DTMark

    It’s hard to believe that here in 2014 we’re still discussing downstream only speeds that are only just into double-digits. It really does make the suitability of BT’s kit for a national broadband network look laughable as Virgin Media’s lowest tier moves up to 50Mbps. I wait for someone to argue that BT has wider coverage and various obligations, but to call yourself a telecoms company I think you’d have to deliver, er, telecommunications suited to a modern world.

    I’m not sure why the copper needs to be ripped out, presumably the E-sides can be removed either by BT or perhaps a scrap metal dealer one day, but then surely that’s easy to replace with fibre and some sort of converter box at the cabinets. This solution must surely have been analysed in detail and planned by now. Surely? With cable, doesn’t the copper only go back as far as the very local cab, which is why cable lines sound bright and clear and BT lines sound quiet and sometimes fuzzy?

    Looking at the D-sides, I think we all pretty well know that a new “last mile” is going to have to be cut in to pavements on a national scale; that BDUK so far is just “throw-away tech”, BT are terrified of going anywhere near those. The copper and aliminium could just be left to rot away and die underground, how much of the underground D-sides are really “recoverable” in any useful sense?

    • Avatar Steve Jones

      I don’t literally mean rip it out, although it would make sense in some circumstances to free up duct capacity or recover the copper (there’s about £2.5-3bn of the metal in the network at current market rates). Of course, the costs of recovery and purification are non-trivial, so there’s significant doubt it would raise much money. What it does mean is withdraw the copper service as to maintain two sets of infrastructure at the same time inevitably raises costs.

      As for fibre to the E-side being (relatively) easily changed for fibre, then that’s precisely what FTTC does apart, that is from the voice. It may be that the FTTC cabinets can also deal with voice services. I would expect that plenty of spare fibre was laid for the FTTC rollout, and there will be plenty of spare sub-ducts for more.

      The problem is the D side, which is where 80-90% of the costs of fibre deployment is incurred. It’s the labour intensive bit of getting fibre to each building. Whether that’s stringing stuff along poles, pulling it down ducts (which are often full or blocked), digging up pavements (the stuff outside my house is direct buries circa 1970) and then runs under a concrete drive).

      It’s not rocket science, it’s money, and logistics and a lot of labour. The cable companies went broke laying co-ax for just 50% of the properties in the country. The consolidated cable business only became marginally viable when the original investors and banks wrote off much of their investment. Quite simply, no investor is going to put tend of billions into a fibre rollout without a prospect of a return. It’s not BT that it has to be sold to. It’s those that will lend money, buy bonds or exercise rights issues. Without that happening, fibre rollouts will be piecemeal.

    • Avatar No Clue

      “It’s hard to believe that here in 2014 we’re still discussing downstream only speeds that are only just into double-digits.”

      Indeed looking at the graph only around 50% have greater than 10Mb, which means 10-15(ish) Million premises are still stuck with less than 10Mb (based on there being 20-30 Million premises in the UK). Clearly the government investment was a waste and the tax payer has been ripped off.

    • Avatar DTMark

      Thanks Steve.

      Your comments make me think of two key points that I’ve made on here for a long time:

      1. The correct strategy is to maximise the amount of private investment that can be levied. The best way to do that is to involve multiple players leading ultimately to co-operation (where the structure can “naturally” support that and lends itself to it, in a way that the current setup does not and can not) and/or competition at various levels from retail through to infra;

      2. That the government’s place in all of this is to act in the capacity of facilitator and that the State not only might, but must, be involved in the cutting in of that new ducted network to make it possible for that to happen. To assist and facilitate using some joined-up thinking, not to hinder or seek to profit.

      .. and..

      That the two biggest failings in the BDUK approach were and are:

      1. Minimising private investment by minimising players;

      2. By doing the latter, severely damaging the opportunity for future private investment, natural competition and the development of markets.

  9. Avatar GNewton

    “That the two biggest failings in the BDUK approach were and are:

    1. Minimising private investment by minimising players;

    2. By doing the latter, severely damaging the opportunity for future private investment, natural competition and the development of markets.”

    Agredd, but I think the biggest failing is the fact that it uses a dubious gap-funding model, which will only burden the taxpayers, and which will mainly only benefit one company (BT) which never had any need for taxpayer’s money. Another failure is the fact that in most cases there is no fibre REPLACING old copper, rather developing and maintaining multiple parallel networks (e.g ADSL-flavours, including LLU, coax-cable networks, VDSL, in addition to POTS over copper rather than VOIP). Most of the taxpayer’s money was wasted on non-future-proof network technologies. And it will only be a question of time when beggars like BT will ask for more public money to do new networks, e.g. fibre.

    On another note: it amazes me to see that none of the utilities companies have taken up the task to build a netxgen network, the ductings, infrastructures, lampposts etc are already there and could easily be used for telecom services by adding fibre, see e.g. https://www.6ginternet.com/fast-broadband-technology

    • Avatar TheFacts

      What ducts do each utility have that could be used for FTTP? Remember Bournemouth.

    • Avatar GNewton

      Just another stupid question here? Have you read the web link I provided as an example?

      BTW.: Ever done your homework? To answer your questions posted some months ago?

    • Avatar TheFacts

      To replace all copper brings back the £25B figure again and would probably take 10 years. Would this see fibre to the property installed in VM areas?

      None of the utilities, gas, electric, water or sewerage have existing infrastructure that would significantly help roll out a FTTP (not wireless) scheme. Or do they? And they would need a significant take up to make the numbers work.

  10. Avatar GNewton

    “None of the utilities, gas, electric, water or sewerage have existing infrastructure that would significantly help roll out a FTTP (not wireless) scheme. Or do they?”

    If instead of only asking question all the times, how about you did some of your own research? I provided you with a link above as a starting point. Also, have you finally done some research on the questions raised some months ago? Have you even tried?

    Here is a question for you: Where did you get the £25Billion figure? And the £2.5Billion BT copper value figure?

Comments RSS Feed

Javascript must be enabled to post (most browsers do this automatically)

Privacy Notice: Please note that news comments are anonymous, which means that we do NOT require you to enter any real personal details to post a message. By clicking to submit a post you agree to storing your comment content, display name, IP, email and / or website details in our database, for as long as the post remains live.

Only the submitted name and comment will be displayed in public, while the rest will be kept private (we will never share this outside of ISPreview, regardless of whether the data is real or fake). This comment system uses submitted IP, email and website address data to spot abuse and spammers. All data is transferred via an encrypted (https secure) session.

NOTE 1: Sometimes your comment might not appear immediately due to site cache (this is cleared every few hours) or it may be caught by automated moderation / anti-spam.

NOTE 2: Comments that break our rules, spam, troll or post via known fake IP/proxy servers may be blocked or removed.
Cheapest Superfast ISPs
  • Hyperoptic £21.00 (*25.00)
    Avg. Speed 50Mbps, Unlimited
    Gift: £50 Shopping Voucher
  • SSE £22.00
    Avg. Speed 35Mbps, Unlimited (FUP)
    Gift: None
  • xln telecom £22.74 (*47.94)
    Avg. Speed 66Mbps, Unlimited (FUP)
    Gift: None
  • Post Office £22.90 (*37.00)
    Avg. Speed 38Mbps, Unlimited
    Gift: None
  • Direct Save Telecom £22.95 (*29.95)
    Avg. Speed 35Mbps, Unlimited
    Gift: None
Prices inc. Line Rental | View All
The Top 20 Category Tags
  1. BT (2553)
  2. FTTP (2291)
  3. FTTC (1687)
  4. Building Digital UK (1626)
  5. Politics (1456)
  6. Openreach (1444)
  7. Business (1273)
  8. FTTH (1124)
  9. Statistics (1118)
  10. Mobile Broadband (1067)
  11. Fibre Optic (984)
  12. Ofcom Regulation (930)
  13. Wireless Internet (929)
  14. 4G (925)
  15. Virgin Media (876)
  16. EE (607)
  17. Sky Broadband (602)
  18. TalkTalk (587)
  19. Vodafone (542)
  20. 3G (418)
Promotion
Helpful ISP Guides and Tips
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
»
Sponsored

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