» Editorial Article » 

UPD Virgin Media UK Eye Future DOCSIS 3.1 Support and 10Gbps Broadband

Monday, December 10th, 2012 (1:09 am) - Score 8,909

The United Kingdom’s national cable operator, Virgin Media, has confirmed to ISPreview.co.uk that the still in-development DOCSIS 3.1 (DOCSISNG) standard, which could one day deliver cable-based broadband ISP speeds of up to 10Gbps (Gigabits per second), is something they’re testing for the future. But what does it mean?

Virgin’s current cable platform is somewhat of a hybrid network that installs fibre to a local node (FTTN) and then delivers the final connection to homes via a short run of copper or high-grade coaxial cable (HFC). The network is largely based off a EuroDOCSIS variant of the ‘Data Over Cable Service Interface Specification’ (DOCSIS3).

In theory the current network gives Virgin Media room to push their speeds up to 200Mbps or possibly even 400Mbps if 8 downstream channels are bonded, although the extra capacity costs, lack of clear demand and possible negative impact on other services that achieving 400Mbps could have might not make it viable just yet. But a boost to 200Mbps or faster is expected. In the meantime customers are already being offered packages of up to 120Mbps.

But Virgin has also found ways to manipulate its platform to deliver trial download speeds of up to 1.5Gbps (150Mbps upload) to businesses in East London, which uses the same network as its domestic customers (here). Last year another cable network adopted a similar setup to demo a 4.5Gbps connection, although this required over 128 DOCSIS downstream channels into a single fibre node. Neither solution would be viable for home users but they do show that cable networks don’t have to switch to a fully fibre optic (FTTH) network just yet.

A VirginMedia Spokesperson told ISPreview.co.uk:

We continue to test various speeds to be ready for when we think the time is right to pre-empt consumer demand and maintain our lead as the UK’s fastest widely available broadband company.”

However, despite DOCSIS 3.0 being established all the way back in 2006, cable standards have not been standing still.

What is DOCSIS 3.1 (DOCSIS NG)?

In October 2012 CableLabs finally unveiled their plans for the new DOCSIS 3.1 specification, which could eventually push cable download speeds all the way up to the dizzy heights of 10Gbps (1Gbps uploads).

The new standard aims to achieve this by harnessing the power of Orthogonal Frequency Division Multiplexing (OFDM), which has been used on all sorts of networks to split a single signal into multiple frequencies (each of which can carry data), and improved error correction (Low Density Parity Check). On top of that cable networks will also need to boost their radio frequency spectrum allocations.

Early estimates suggest that such enhancements could add as much as 50% extra capacity to cable networks and would still be backwards-compatible with existing DOCSIS gear (CPE), which is a vitally important consideration and would help to keep upgrade costs under control (CPE changes are often one of the most complicated elements to update).

Timetable for DOCSIS 3.1 from VirginMedia

A Virgin Media spokesperson confirmed to ISPreview.co.uk that they “could theoretically integrate this into our existing network” and without too much trouble thanks to its backwards compatibility. But the specifications are still under development and as a result Virgin are currently only experimenting with such solutions as “an exercise … rather than a trial“.

The spokesperson added that Virgin Media expected the upgrade costs for such work to be “relatively minor” in contrast to the “many billions needed to create BT’s FTTC network“. It should however be noted that BT’s network covers most of the UK and it was upgrading from a pure copper platform (Virgin reaches about half of the country.. mostly in urban areas), thus the comparison isn’t entirely balanced. Virgin added that moving to DOCSIS 3.1 would require a similar effort to the £110m they’ve already spent on doubling speeds for over 4 million of their customers.

Meanwhile the current CableLabs time-table for DOCSIS 3.1 development suggests that the preliminary specification should be ready within the first half of 2013, while the first trial kit could be ready for proper testing in early 2014. Sadly few expect the final hardware to be available before 2015.

At this stage it’s still far too early to know what home consumers could expect to get from the platform update, although it’s not unreasonable to predict that a 400Mbps+ service under DOCSIS 3.1 would be very viable and without an excessive upgrade cost, relatively speaking. Delivering affordable speeds like that would put a lot of pressure on BT’s slower FTTC service, especially with the faster FTTP-On-Demand alternative set to be quite expensive (installation fee).

In theory Virgin’s home connectivity under DOCSIS 3.1 could eventually approach 600-800Mbps (the 10Gbps figure would only be viable for business users) but that’s just speculation until we see how the new standard works in a real-world environment. On top of that Virgin’s recent peering and congestion problems show that upgrading the infrastructure is only ever half of the battle.

UPDATE 7th March 2013

Virgin Media’s network man, Phil Oakley, and Ralph Brown, CTO of CableLabs, have told the Cable Congress 2013 conference in London that broadband download speeds of 1Gbps could in theory arrive in homes by 2015 or 2016 (much as we first suggested above).

Share with Twitter
Share with Linkedin
Share with Facebook
Share with Reddit
Share with Pinterest
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
22 Responses
  1. FibreFred says:

    “But what does it mean?”

    More congestion for VM customers?

  2. Ignitionnet says:

    No more so than the move to 17a meant more congestion for FTTC customers. More capacity per MHz on the cable network really making higher speeds a possibility.

    How VM plan the capacity will decide congestion or not.

    1. FibreFred says:

      They haven’t capacity planned anything with their upgrades so far hence my comment

      Why is it no more so, have you seen FTTC customers complaining in their masses about congestion like you see on the VM forums?

      I’m saying the VM network in places is already congested, oversubscribed OBR’s and local congestion. It has been a problem for a long time and still is AFAIK, increasing the headline speeds only compounds the problem further.

    2. DTMark says:

      “have you seen FTTC customers complaining in their masses about congestion like you see on the VM forums?”

      Yes, actually, I have. People whose “up to” 80Meg FTTC service only runs at about 25Meg anyway because of the ancient wire used to deploy it so it doesn’t even have “burst” capability, who then see slowdowns at various times in particular in the evenings, and generally sporadic performance. The problems seem to be localised to specific exchanges.

    3. Anoyed tax payer says:

      @DTMark The slow down from 80Meg to 25Meg you mention is not related to congestion

    4. FibreFred says:


      Not so sure about that

      Type virgin media congestion and BT FTTC congestion into Google and see the difference

      And as pointed out ancient wires are not related to congestion

      Virgin have had problems with their network in areas for some time now, this talk of higher headline speeds will make matters worse. Better to invest in capacity and make your customer happy before they leave

    5. DTMark says:

      I did say “and then falls further” (from the 25Meg). FTTC is new for BT, so it’s early days yet.

      Try searching for “BT Infinity congestion”. If congestion was not actually expected, there would not be prioritisation in place on the network. Also remember that the minimum committed sync rate on Infinity is nil / 5 Meg / 15 meg, and that’s not even throughput.

      Given every cable broadband customer “syncs” at the full rate, congestion is fairly easy to spot, though I suspect some reported “congestion” is in fact failing power levels which is a technical issue which should be solved.

      It’s harder to say with FTTC whether it’s congestion, crosstalk (which has been demonstrably shown to shave chunks off peoples’ max attainable rates, sometimes quite significantly) or a technical fault which could be resolved since BT decided to lock the users out of getting data from the modems.

      That said: that the competition is still trying to shove data signals down phone wires with sometimes very poor results is no excuse for poor network planning or overselling.

    6. FibreFred says:

      Spotting congestion on Virgin Cable isn’t a problem its getting Virgin to do something about it that is the problem as many of their customers will testify

      I’m not saying there isn’t congestion anywhere on FTTC, I’m saying its definitely a problem for Virgin which they’ve yet to fix


      So where is all this congestion you speak of with FTTC?

    7. FibreFred says:

      Here’s a very typical story and why Virgin need to invest in improving what they have now before upping their speeds again


    8. Ignitionnet says:

      Congestion affects the two platforms differently DOCSIS handles congestion somewhat less gracefully than DSL.

      There is congestion on the BT Wholesale network but it’s relatively unnoticed due to being managed more gracefully. It’s congestion at backhaul level rather than access network level as with DOCSIS so can be groomed more elegantly.

  3. DTMark says:

    “Keep up” indeed 😉 Look what happens when you invest in a modern network. Competition raises standards. And, no BDUK money! How can this be possible? 😉

    Do we have any more information about what the c. £110 million spend is “for”, and how complete it is?

    When I last looked, the 100Meg service was averaging about 85 to 90Meg in real-world performance, but that masks undoubted congestion in certain segments.

    I believed that the money was to upgrade all the cab to cab interlinks which are still old copper co-ax, to fibre, which would then prep the network so as to deliver these sorts of speeds (400Meg+) in reality.

  4. zemadeiran says:

    The only real way for Virgin Media to sort this shit out would be to migrate their users to FTTH via their VM cab.

    The rest is pissing into the wind…

    1. Ignitionnet says:

      Sadly most VM cabs don’t have fibre to them.

  5. FibreFred says:

    ” Delivering affordable speeds like that would put a lot of pressure on BT’s slower FTTC service, especially with the faster FTTP-On-Demand alternative set to be quite expensive (installation fee).”

    I really don’t see it myself. Not without Virgin spending a vast amount of money overhauling their own network. For me it seems that Virgin are stuck in a loop:-

    1) Advertise amazing speeds and get most of the ad’s pulled over time after the ASA tells them off
    2) Invest small change every few years to increase headline speeds but put more pressure on the local network and core (UBR’s)
    3) Ignore those customers that complain about congestion and carry on looking for new sign ups to compound the issue further
    4) Go to 1

    I mean if they really did spend money (I know they are in debt) they really could make it an amazing network, fibre to each cab and GPON the lot, replace (over time) the coax to the home with fibre etc etc. It could be the best, sadly I just don’t see it happening, the just seem content with relatively cheap tweaks which just makes matters worse

    1. Ignitionnet says:

      That you describe uBRs as ‘core’ makes me question your expertise in this matter sir.

    2. FibreFred says:

      Ok uplink to the core, its a pinch point, an aggregation of traffic from the streets to the core network

    3. DTMark says:

      It was my impression that the £110 million spend (the “resegmentation project”) was about “fibre to every cab” e.g. replacing co-ax interlinks.

      Given that’s very labour intensive you’d have thought plenty of capacity would be being provisioned, so as to only do this once a decade or less.

    4. Ignitionnet says:

      The cash was for line cards, access network upgrades and node splits. It’s not feasible to put fibre into every cabinet for a few reasons, a big one being that most cabinets are line powered, another being that some network architectures on the cable network don’t even use cabinets for the most part but are a similar architecture to many USA networks with a coaxial trunk line down the street which is split to feed home drops periodically and amplified to compensate for loss periodically.

      I know what you’re talking about, fibre-deep architecture where everything bar the line serving customers is fibre and there’s at most a line extender (in-line amplifier) and no amplified splitting of cable, but that would’ve cost a fair amount more than VM have spent.

      VM will get to the fibre-deep architecture eventually though of course, just a step at a time rather than in one burst in a similar manner to how BT are deploying FTTC in most cases, pushing fibre deeper towards homes, rather than going straight to FTTP. VM are in the position where they can do this in more stages than BT who realistically only have 3 places where fibre can go into their loops, not at all, to PCP, or to premises.

      It’s a bit disingenuous, by the way, to refer to VM’s upgrades as ‘minor tweaks’. VM’s network is a collection of smaller networks and some of them were built as cheaply as possible with the result that for VM to deliver 10:1 upstream ratio they had to replace most of the powered components in the networks and in some areas replace many unpowered components due to issues. In my own area for example many amplifiers, fibre nodes and optics both in hubsite and in the field had to be replaced as there was only a 25MHz range available for upstream traffic and less than 500MHz for downstream. Remember this is shared between the entire MAC domain and at a maximum of 4 bits per symbol VM use at the moment and accounting for some parts of the spectrum being unusable due to noise the 25MHz didn’t really cut it, not that the lasers in the area could’ve coped with the loading upstream anyway they’d have been overdriven and clipped like crazy being old Fabry-Perot diodes originally installed to deal with 800kHz QPSK return channels for interactive services not multiple 6.4MHz 16QAM channels with some 3.2MHz 16QAM on the side for non-DOCSIS 2 CPE.

  6. Ignitionnet says:

    The CMTS themselves are fine, the issue is individual MAC domains, these issues are resolvable through reducing the size of MAC domains and/or introducing more channels.

    Zero need to deploy fibre to each cabinet just yet, that’s only needed to compete with FTTP, DOCSIS itself is perfectly capable of robust services at 200Mb down and 20Mb up and VM could hit that level of performance with these same relatively minor tweaks once the network was properly segmented – 8 downstream channels, 4 upstream channels, sub-250 homes passed MAC domains.

    Absolutely no need to deploy FTTP or indeed FTTC at all, the existing FTTN architecture is fine, with the nodes getting smaller physically and/or via DWDM solutions.

    1. babis3g says:


  7. Phil says:

    the wireless will not be able to cope with 200Mbps + anyway, seem pointless to have more faster speed. 80 Meg is faster enough these day.

    1. Corrado Mella says:

      You obviously live on your own… Lucky you.

      Typical household, a family of 4 can have a device or two each (mobile, PC/laptop, tablet…) and with more “communal” devices – wired or wireless – like Games Consoles, Catch-up TV boxes, internet radios, etc… even on an aging 802.11g wireless network it means a dozen or so devices can stretch a 120Mbps cable BB (delivering true 120Mpbs!), imagine that on a 80Mpbs (in theory, more likely 55-60) FTTC.

      It’s not the quantity of data you download in time, but the instant availability of bandwidth to service all devices that require it, that is becoming the most contentious issue. When you click on a file on DropBox, or a song on your iCloud, you want it there and then, so you need instant speed bursts rather than sustained traffic volume.

      That’s why Virgin is mostly investing in speeds to the CPE, and not that much in the core network. Where the network is oversubscribed and congested, is both a case of historic oversubscription from legacy providers acquired years ago and – sadly – piracy on the DOCSIS1.1 network that can’t be shut down as it’s used by the TIVO boxes.

      There’s no immediate solution to the congested areas but resegmenting the network. Here north of the border, in a relatively uncongested network area, it’s happened three times in 8 years. Give them time and service will get to a good standard everywhere.

Comments are closed.

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
  • Vodafone £19.50 (*22.50)
    Speed 38Mbps, Unlimited
    Gift: None
  • NOW £20.00 (*32.00)
    Speed 36Mbps, Unlimited
    Gift: None
  • Hyperoptic £20.00 (*25.00)
    Speed 50Mbps, Unlimited
    Gift: Promo Code: HYPERFALL21
  • Shell Energy £21.99 (*30.99)
    Speed 35Mbps, Unlimited
    Gift: None
  • Plusnet £22.00 (*38.20)
    Speed 36Mbps, Unlimited
    Gift: £70 Reward Card
Large Availability | View All
New Forum Topics
Cheapest Ultrafast ISPs
  • Gigaclear £24.00 (*49.00)
    Speed: 300Mbps, Unlimited
    Gift: None
  • Vodafone £24.00 (*27.00)
    Speed: 100Mbps, Unlimited
    Gift: None
  • Community Fibre £25.00 (*27.50)
    Speed: 200Mbps, Unlimited
    Gift: None
  • Hyperoptic £25.00 (*35.00)
    Speed: 150Mbps, Unlimited
    Gift: Promo Code: HYPERFALL21
  • Virgin Media £28.00 (*52.00)
    Speed: 108Mbps, Unlimited
    Gift: None
Large Availability | View All
The Top 20 Category Tags
  1. FTTP (3570)
  2. BT (3024)
  3. Politics (1943)
  4. Building Digital UK (1929)
  5. FTTC (1888)
  6. Openreach (1838)
  7. Business (1694)
  8. Mobile Broadband (1480)
  9. Statistics (1410)
  10. FTTH (1365)
  11. 4G (1278)
  12. Fibre Optic (1176)
  13. Virgin Media (1174)
  14. Wireless Internet (1163)
  15. Ofcom Regulation (1150)
  16. Vodafone (847)
  17. EE (837)
  18. 5G (772)
  19. TalkTalk (770)
  20. Sky Broadband (748)
Helpful ISP Guides and Tips

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