Following the broadband money

Why 19Mbps won’t be enough

with 27 comments

The present high speed broadband roll-out will see the UK lose its status as a first world economy, contends Wispa CEO Richard Brown.

Brown plans to challenge the Broadband Stakeholders’ Group conclusion that the average UK broadband user will not need more than 19Mbps download speed by 2023.

His talk to the Chartered Institute for IT (formerly the British Computer Society) in Wales will be webcast on Wednesday, 29 January at 18.00 on YouTube, while his presentation is available without commentary on Prezi.
Brown plans to examine the effect the BSG’s endorsement of BT’s fibre to the cabinet (FTTC) strategy in the light of competitive ubiquitous broadband roll-outs and historical UK investments in ubiquitous infrastructures.

Brown has been a vociferous critic of the government’s broadband contracts, arguing that done differently, it would be possible to deliver 100Mbps to all for less than the £1.4bn of taxpayers’ money that governments are giving BT.

“Predicting usage is pointless; enabling usage is essential,” he says. “Does it really matter how? No, but it really matters if.”

One of the killer apps is ultrahigh definition TV or 4k TV. Exactly two years ago BT was telling housing developers in-home networks would need to run at around 250Mbps to support 4k content cached in a local recorder. At this month’s CES International show, Netflix boss Reed Hastings suggested that a new codec, HEVC, would allow 4k to be streamed over a 15Mbps in-home such as Wi-Fi.

But getting the content into the house is going to be the problem, something content distribution network operator Akamai is looking at. Basically it means caching content closer to the end user, and that might mean giving Akamai access to your various devices so that it can pre-load content for you.

But what about live streaming the Olympics in 4k? That’s when the copper and air links between you and the core network are going to resemble a human trying to pass a kidney stone. And as painful.


Written by Br0kenTeleph0n3

2014/01/28 at 06:58

Posted in Uncategorized

27 Responses

Subscribe to comments with RSS.

  1. Do you even know what 4k needs interms of bandwidth?


    2014/01/28 at 07:26

    • @Neil,
      I know that my 18.2 Mbps (sync rate) ADSL2+ circuit often struggles to stream iPlayer SD today. I hope that Mr Brown also addresses the necessary change in focus by all ISPs away from a speed obsession to Quality of Service. I’m not aware of any BT work to improve on that, or towards changing their commercial model towards a QoS basis rather than “advertised/lied-about/theoretical” bandwidths which are impossible to predict or maintain with accuracy on a rural DSL line – I guess the money is good the way things are for BT.

      The “future” technologies they plan on deploying over the copper don’t talk about commercially viable QoS either. Perhaps you could share some insight to convince me otherwise? Surely it is in BTs interest to maintain the dreary copper for a few good years more by introducing such a QoS focused model?

      According to Netflix, who are in the business, a 4k stream currently requires 15.6 Mbps with, I expect, a very, very, very low error bit-rate. Neither BT nor I can comment on future (not yet invented) codec efficiency, however historical codec achievements suggest digital signal transmission codecs don’t improve anywhere near as impressively as the silicon transistor density has broadly achieved according to Moore’s Law.

      • sorry but that article looks inaccurate in my view and every piece of work we are doing in BT is focused on quality of service rather than top line speed, including a complete replacement of the 21C network over the past 2 years.

        a huge amount of traffic delivered in BT’s retail network is served from within the BT network as opposed to other networks. we have a long way to 4k being mainstream (I have a 4k platform at home)

        if you are having challenges streaming iPlayer at that rate something else is going wrong –
        I have a 6m line that streams iPlayer perfectly fine.

        if you want I can look into your line if it’s a BT line email me your phone number to my BT email which is first name dot last name at BT dot com

        Neil McRae


        2014/01/28 at 09:19

    • I think it changes with the tech. Until H.265 was standardised, which halves the bandwidth you need for a high quality transmission, you needed more. Since BT was saying in January 2012 we need 250Mbps for in-home cached transmissions, what’s your present estimate for live streaming?


      2014/01/28 at 10:18

  2. Also where is this magic business case ? Surely anyone making these bold statements should have a case and a model?

    Hmm and look at the network development going on inside BT! How coincidental!


    2014/01/28 at 07:36

    • ‘Magic business case’ for what?


      2014/01/28 at 08:54

    • You might want to ask the at least 22 European competitors ahead of us in FTTH penetration to borrow their business case:

      Companies and individuals involved in Google’s 1TB experiment don’t seem to be having any difficulty in demonstrating incremental benefit, either.

      By the time you’ve “proved” the benefit – and how many billion-dollar internet businesses would ever have got funding on the basis of a BT-strength business case? – they’ll be even further ahead of us. I’m sure we’ll all be very grateful for BT and HMG for squandering for the UK’s #1 spot in global ecommerce.

      There aren’t many industries in hypergrowth mode where the UK claim such a leading position. It beggars belief that the government and BT aren’t bridling the economy with the absolute best available infrastructure. The mobile internet network is in a shocking state too, still inaccessible even in vast areas of central London.

      Once the leading country for rail in the world, the UK might get HS2 some 50 years after TGV and 65 years after the bullet train. It’s great to see the myopic UK monoliths learning from the past.


      2014/01/28 at 09:09

      • Trouble with those cases is that the benefits apply to a wider distribution AND can be delivered via FTTC. I’d love to see fibre everywhere but I also want to ensure that what we give better connectivity to as many people as possible.


        2014/01/28 at 10:11

      • Which is why ubiquity is so important…


        2014/01/28 at 10:34

  3. Brown has been a vociferous critic of the government’s broadband contracts, arguing that done differently, it would be possible to deliver 100Mbps to all for less than the £1.4bn of taxpayers’ money that governments are giving BT.

    the above, I’d like to see that done in my local town of harrow. if it was so easy then why isn’t being done ?


    2014/01/28 at 09:24

    • The white paper that Ian refers to is mentioned in the presentation – (includes grosse costing information, functional understanding of mechanisms, and expansion of policy interaction with economy) this isn’t micro delivery, but macro delivery, pan UK. If you’d like to spend some time taking a look, you can find a copy at where you are also free to download a copy.


      2014/01/28 at 09:53

    • As for why it hasn’t been done…..


      2014/01/28 at 09:53

      • i am sure it has nothing to do with retaining the copper (or aluminium) investment and promoting repeated state funding every 5-10 years for the next small concession?

  4. If you are in or around Newtown in mid-Wales tomorrow, come along in person, its 6pm in Theater Hafen.

    Clive King

    2014/01/28 at 09:30

  5. Anyone able to watch this?


    2014/01/29 at 18:05

    • No. They can’t get a connection. Superfast Cymru is conspicuous by its absence.


      2014/01/29 at 21:25

  6. Presentation with commentary now here:


    2014/01/30 at 22:37

  7. Apparently a technology failure with the recording, hence redone.

    Richards talks about the 273,731 number and difference between premises ‘passed’ and ‘delivered’. However DCMS say:

    These premises numbers therefore exclude premises which benefit from the BDUK-supported projects if their resulting speeds are below 24 Mbit/s, even if these benefits help meet BDUK’s Universal Service Commitment to raise broadband speeds above 2 Mbit/s in project areas not achieving superfast speeds. They also exclude overspill effects of BDUK-supported projects on premises which already have superfast broadband available.

    So the numbers exclude those with a long distance from the cabinet.

    ps I can recommend the talks given by the various chartered institutes round the UK, usually free with tea/coffee/sandwiches.


    2014/01/31 at 09:59

    • Peter – I appreciate the comment – your interpretation is different to mine.

      I’ll explain why.

      You state that long distance from cabinet is excluded. It isn’t . It isn’t, because each of those 273,731 premises have not yet been tested as to their capability to achieve superfast speeds.

      Long distance
      Poor quality copper
      No existing line
      Aluminium repairs
      Poor quality junctions

      Are all yet to be tested/surveyed/discovered.

      A property can be extremely adjacent to a cabinet and yet have hundreds (maybe a thousand or more) km of cable run to the property due to the nature of the poles and ducts. When the poles and ducts were installed it was not critical to keep the runs short and so they ran as conveniently as possible.

      In London there are lots of complaints that (paraphrasing) say “I can see the blimmin cabinet, but can’t get superfast”.

      Additionally, as mentioned in the presentation – Clive had an Openreach survey on the day I spoke to the Institute. The nearest that Openreach could get was a telegraph pole just under 1/2mile away. He is now negotiating as to how many thousands of pounds it is going to cost to get FTTP, as it is not possible to serve him with superfast speeds with FTTC over the existing copper and so a new run/duct will be required.

      I want this to work. I know how desperately the UK needs this to work. It isn’t working – and misleading you, and others by not quite meeting the requirements for openess isn’t helping us to make it work.

      I nicked the following from here: for convenience as Andrew has already laid it out in table form

      Distance Download Upload
      100m 100 Mbps 25 Mbps
      150m 80 Mbps 20 Mbps
      200m 65 Mbps 18 Mbps
      300m 45 Mbps 17 Mbps
      400m 42 Mbps 16 Mbps
      500m 38 Mbps 15 Mbps
      600m 35 Mbps 14 Mbps
      700m 32 Mbps 11 Mbps
      800m 28 Mbps 10 Mbps
      900m 25 Mbps 9 Mbps
      1000m 24 Mbps 8 Mbps
      1250m 17 Mbps 5 Mbps
      1500m 15 Mbps 4 Mbps

      This is only a rough guide – but as you can see, a property is going to need to be within 1km to meet the Gov requirement of superfast and yet Clive’s property is 800m or so from the nearest access and it is going to cost him if he wants superfast.

      It really isn’t adequate to assume premises passed aren’t premises passed by, regardless of the assumptions you are handed.


      2014/01/31 at 11:01

      • I’m just copying the text from DCMS which does exclude properties clearly not able to get superfast and there will be those additional due to other reasons. Question is how to count them.

        In the spirit of openness, who are you?!


        2014/01/31 at 11:10

    • @PeterBarr Thank you for the BDUK statement.

      BDUK should tell local authorities of their policy not to count premises with speeds below 24Mbps so that they all sing from the same hymn sheet.

      Currently it seems that they all do as Surrey County Council and include all promises connected to a partially enabled fibre cabinet in their statistics. This disregards those that cannot get greater than 24Mbps and those that will have to wait for further installation of ducts, tie lines and additional cabinets once the initial installed capacity fills up. Does anyone know who will pay for this extra infrastructure as it is likely that the demand that busts the initial capacity will not happen until after the local authority contracts complete – no doubt with much publicity on how they have provided superfast broadband to more than 95% of premises.

      David Cooper

      2014/01/31 at 22:27

      • Perhaps they can use the 30% saving on overhead costs that BDUK is identifying in BDUK contracts to fill in the missing bits.


        2014/01/31 at 23:24

  8. Richard Brown
    I did the presentation.

    Not entirely sure how to really answer that given that we only have your name Peter so I aimed for just stating sufficient to establish why I am commenting…

    Copying text from websites/releases etc is pretty much the problem, however – your assertion that it does not include properties ‘clearly not able to get superfast’ is (unfortunately) not as clear as you may think. As I mentioned previously – properties that should be able to get superfast (according to known measurement factors) are often simply not able to. This is the fundamental issue with premises passed.

    According to Welsh Gov press releases Superfast Cymru will deliver 96% of the population with superfast – the contract is 95% of the population with premises passed.
    Press releases and website for Welsh Gov state £220m to be spent – contract shows £195m
    ad infinitum

    You are on a blog that doesn’t act as BT/Gov apologist and doesn’t simply reprint the Press releases – I would have thought you would be keen as mustard to make sure that what you are copying is accurate before you paste it?


    2014/01/31 at 11:32

    • JIC anyone hasn’t twigged, Peter Barr is Peter Barrington, aka Somerset. In the spirit of openness.


      2014/01/31 at 13:14

      • LOL – well, that explains the requirement for extended explanation. TY


        2014/01/31 at 16:14

  9. This is a good topic for

    More or Less: Behind the Stats – Tim Harford investigates numbers in the news. Numbers are used in every area of public debate. But are they always reliable?


    2014/01/31 at 17:20

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: