It's still a form of FTTx. Fibre to the Node

In reply to a post by kwikbreaks:

The confusion over what is fibre broadband was started by VM marketing and was intended to confuse - it sounds like it confused their staff as well.

I regard this sort of disgraceful usurping of technical reality for marketing purposes as unacceptable - but it seems that the ASA is now content for any broadband service making use of fibre to be marketed as **** FIBRE BROADBAND **** (with far more bells and whistles than is possible on a textual forum).

I'm waiting for someone to market dial-up as fibre Internet access, on the basis that the ports on the peering switch are almost certainly some form of fibre Ethernet.


Virgin is a two-way hybrid fibre-coax network - they've either upgraded, closed (Westminster) or stopped marketing (Milton Keynes) the areas not capable of two-way HFC. Gradually, fibre has got deeper into their network, but possibly is not as geographically close to the customer as with BT Openreach FTTC deployment.


The two key differences between Virgin and BT's FTTC deployment (other than the different final hop technology) are:

* BT are required to allow third-parties wholesale use of their network on the same basis as their in-house retail business

* BT's deployment makes provision for FTTP build-out using spare fibre deployed alongside the FTTC build (possibly on BT's initiative, but in the medium term more likely as FTTP On Demand when that launches).


Ultimately, the precise technology used matters little to the customer experience - even to someone like me who used to work in computer networking R&D. All the posts about the alleged superiority of coax or twisted pair are really fanboy/fangirl stuff. What matters is what speed the technology is capable of, the contention experienced, whether a particular service is available to you and whether the network offers wholesale capacity (giving you a greater choice of providers).


We have several VM TiVo boxes, but do not regard VM broadband as suitable for our home business use. We had a VM cable modem in the early days, and the contention was awful, though I have no idea what the current situation is like locally now (probably not good, as VM offered cable modem service here long before BT enabled the exchange for ADSL, so the majority of homes are probably still on VM).

The terms and conditions of VM's residential service used to prohibit business use - though I have no idea how they'd distinguish.

We have a routed /28 with Zen - to get anything vaguely similar on VM would involve taking an overpriced business broadband product.


Fortunately, BT switched on FTTC in this area a few weeks ago and our Zen connection will be upgraded to FTTC in early January. We're only around 200m from the cabinet in a relatively modern estate with underground ductwork, so an eventual FTTP upgrade will likely not be too difficult. That said, we're not likely to be queueing up to pay for FTTP On Demand when it launches!

There is a place for VM in the residential Internet access marketplace - but it would be so good if all the providers would adopt a 'truth in marketing' standpoint. Unfortunately, I realise this is as unlikely as flying porcines.


This post is brought to you by what these rotten marketers would describe as a fibre LAN. The servers here are in an air-conditioned room in the garage, with the house switch and server rack switch connected to each other by two 1000BaseSX gigabit fibre links. Of course, the rest of the hardware is connected by gigabit twisted pair or wireless - it's a LAN with a fibre link where it matters for galvanic isolation and regulatory compliance. Using fibre for other links would just be a waste of money (I doubt you'll get a pair of 1000BaseSX SFP GBICs for under £100).

In reply to a post by Stanman_24:

other side of the coin is how else to describe their broadband ?

partial fibre broadband

FTTC Bradband


Fibe to your cabinet

The simplest answer is fibre to the node - it isn't FTTC as the majority of cabinets have no fibre feed at all they are simply coax distribution points fed by coax.

What is true is that distance to the cabinet does not impact on the headline speed with cable so your headline speed is whatever you are paying for. The available bandwidth though depends on how well the sales team did in the area and whether the network team managed to keep up.... Typically several hundred VM customers share 200 - 400Mbps maximum pipes with most at the lower end of that range. I have 5 TV channels down so 250Mbps and the area has two non-bonded upstreams (that I've seen) so 2 x 18Mbps up. A single torrent freak operating through a VPN can screw it up for the other couple of hundred or so on the same network segment. Two can just about kill it.

they probably do but thats probably because they under pressure to do so, VM is ran purely for sales and retention.

The agressiveness for sales is unreal, before I opted out I was getting dozens of cold calls a month to my VM landline and still getting the almost daily junk mail through my letterbox.

what i find is more the issue is Upstream overutilization with VM , download speed it self is not the issue, i have found its the packet loss due to only having 4 upstream channels on the Customer Coax side of the FTTN node compared to 110 downstream channels (it makes no sense to offer more upload speed if they are not fixing the node upstream issues, as that will add to the issue)

so once every one gets home and fires up there pc that has utorrent running in the back ground (as its set to start on startup), set to unlimited upload by default it drowns out every one els who wishes to use the service and the end result is packet loss insures

the fix is they need to add more upstream channels and implement QOS at the User modem level to prevent packet loss), as i find that is more an issue then the speed

if you look at my BQM page (and the mouse one) in my sig you can see it happening (high pings then packet loss, Ignore the ones at the bottom as they turn off the pc most of the time i need to remove them)

Edited by leexgx (Thu 13-Dec-12 20:15:41)

yes I am well aware of upstream utilisation issues, I have suffered from it myself, although I didnt know the imbalance was so severe that its just 4 channels to 110 downstream.

are you sure its that much?

because if we assume even 200 users per upstream that would still only be 8 users per downstream which seems low.

All of the channels get reused many times over so there is no real way of knowing what the up/down ratio is at the local level where it actually matters. In my area it's 5 bonded down to 2 non-bonded up so far as I can see. (so 250Mbps down and 2 x 18Mbps up).

It's obvious there can't be 110 downstreams actually in use in any given location as that would just about be the total coax capacity and it carries umpteen TV channels as well.

Edited by kwikbreaks (Thu 13-Dec-12 22:29:51)

In reply to a post by kwikbreaks:

All of the channels get reused many times over so there is no real way of knowing what the up/down ratio is at the local level where it actually matters. In my area it's 5 bonded down to 2 non-bonded up so far as I can see. (so 250Mbps down and 2 x 18Mbps up).

It's obvious there can't be 110 downstreams actually in use in any given location as that would just about be the total coax capacity and it carries umpteen TV channels as well.

Virgin will inevitably use physical segmentation of the network to permit re-use (assuming crosstalk and similar RF considerations allow) of frequencies used by cable modems.


Upstream exhaustion requires careful management in any cable deployment. The necessity of dealing with multiple talkers requires a greater degree of redundancy on the upstream to the downstream, mandating lower speed operation. As well as P2P traffic, increased use of cloud storage, people home-working using VPNs and increasing use of video calls such as Skype means the demand for upstream bandwidth will continue to grow even though residential Internet traffic remains highly asymmetric. Unfortunately, traffic management of a cable upstream can only go so far - it isn't possible to assess the relative importance of the various user's intended traffic when deciding which cable modem transmits next, also traffic management works much better when imposed at the transmitting rather than receiving end of a link.

It is only possible to take physical segmentation so far - fibre and power availability only run so deep in Virgin's network. Ultimately, Virgin have to sell highly contended service to be economic (as the contention is imposed by the shared nature of the network beyond the node) and have taken the understandable marketing decision to focus on the high speeds available across their coverage footprint (a unique selling point compared to DSL). Both these make Virgin's service vulnerable to the effects of heavy users.


When I was on Virgin, my experience is that achievable speed in both directions fell a long way short of your bandwidth cap at peak times - especially in the upstream direction. This was in the days when DOCSIS 1.1 was king - I remember my cap being of the order of 600kbps/100kbps on an 12Mbps/1Mbps network. The ratios between these numbers have remained reasonably similar, even though the network now operates around 25 times faster.


One big advantage of FTTC is that the connections between the DSLAM and the end user are uncontended (though the more users there are, the more crosstalk will lower sync speed for all). The contention comes in the backhaul from the DSLAM to the aggregation point (typically but not always the local phone exchange in the BT system) and in the network beyond. These links can be upgraded relatively cheaply by lighting spare fibres in the existing bundle or deploying technologies such as CWDM and DWDM.


Ultimately, both Virgin's FTTN and BT's FTTC push fibre closer to the end user than traditional ADSL. Both technologies are limited by the presence of some copper (or, in BT's case, sometimes aluminium), but whilst many on this site would welcome a pure fibre connection, the economics still do not make sense in many cases. It is a shame that the marketers have so obfuscated the distinction between 'deeper fibre' technologies like FTTN and FTTC and 'pure fibre' FTTP.

BT's FTTP On Demand product will help to drive deployment of pure fibre connections. The on demand product seems likely to be limited to areas that already have FTTC (as it is likely to use fibre blown in the same tubing as that running to the FTTC DSLAMs) and the one-off build costs levied on customers may well be unaffordable for many. Those who will benefit most are those furthest from the DSLAM who will likely have the most to pay. The difficulty and expense associated with FTTP may help explain why a considerable proportion of BT's proposed commercial FTTP deployment has been delayed or cancelled.

However, BT are actively deploying fibre closer to customers, and offer third parties access to their deployed fibre-based infrastructure. Virgin seem content to make the most out of their already deployed HFC footprint - so far as I am aware, they are not extending that footprint to any significant extent, are not rolling out any residential FTTP and show no interest in offering wholesale access to their infrastructure.

In reply to a post by David_W:

Both these make Virgin's service vulnerable to the effects of heavy users.

I certainly agree with that but it seems VM marketing don't as another feature of their advertising apart from high headline speeds is "unlimited downloads".

IMO the days of cable are numbered and it will be superceded by FTTC/FTTP as end-user bandwidth requirements continue to increase. It may live on as a poor mans broadband but there will certainly be no significant expansion in coverage areas.

Edited by kwikbreaks (Fri 14-Dec-12 09:42:31)