|
|
Eircom started rolling out vectoring in Ireland last spring.
In a recent publication by the Irish government (showing consultation responses to their equivalent of BDUK) , we get to see a graph of speed vs distance. In the published document, see page 97.
How much does this apply to the prospects of vectoring in the UK?
Ireland tends to use 0.5mm copper like us, are using Huawei DSLAMs, and their equivalent of the ANFP puts the same power masking requirements that we have.
However, they run their DSLAMs with a 9dB target margin. If they used 6 dB margins, speeds could be 11 Mbps higher on the left half of the graph, with the increase dropping down to an extra 2 Mbps at the longer distances.
Here's a copy of the graph in that document, with an extra line showing theoretical behaviour with a 6dB margin.
http://i59.tinypic.com/2s68vty.png
I don't see any other significant differences in their line plant. As ever, words of caution need to be added for aluminium lines.
|
|
|
Interesting thanks.
SKY FTTC - 40/10
|
|
|
|
Interesting, It would have been nice to have seen non-vectored on the same chart, However, if I take the approximately 700m line I have (currently synced at 57mbps), then that would forecast around 80mbps.
Quite what level of VDSL uptake this represents would be interesting to know. Even though vectoring is designed to reduce x-talk issues, I doubt the effect is eliminated entirely.
If this works as advertised, it would, of course, considerably extend the reach of "superfast" 24mbps up to around 1.6km/1mile. I would have thought that would have had a material impact on BDUK contracts that have targets based on such coverage (or any other higher speed).
It does make me wonder what the position is with the Openreach trials. Is there some complication we are unaware of?
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
Well thankfully
My line max attainable is 140mbps and my current snr is 26 (or so my ECI says) so will probably not need vectoring i'd hope
i've not encounted any issues with crosstalk yet maybe when a 120/30 service is released and my cab fills up completely.
i'm about 300m ish from the cab.
as trials go i heard from someone unconfirmed that BT would be rolling this out apr next year but ofc grain of salt with this
far as i know the trials have been going well or at least i've not heard of any set backs.
SKY FTTC - 40/10
Edited by epyon (Mon 10-Nov-14 19:32:07)
|
|
|
|
I have to say, I was quite impressed by the results there. I had expected the range of "superfast" to be extended some 200-300 metres, but 1600 metres would be more than I really expected.
And, as you say, interesting implications for BDUK - but equally interesting consequences for the extender technology, like FTTRN. It might alter the planning decisions about where these nodes should go.
There is anecdotal evidence that the rollout seems to have been Huawei-heavy recently. With no visible progress with ECI's vectoring, this might be a reason why.
|
|
|
I think its generous to count openreach's network as 0.5mm copper
its a complete raffle what cable you get.
the last 40m of my line including the drop wire is 0.3 ali. The rest is even thinner copper but I been told by a openreach guy on a forum thats impossible so who knows.
Since this 40m of ali will be with me even on g.fast any estimates I consider hopeful at best.
However on the other hand since crosstalk is worse on ali, maybe ali stands to gain the most from it?
|
|
|
Since this 40m of ali will be with me even on g.fast any estimates I consider hopeful at best.
You expect to live long enough to see G.fast deployed? Go home you're drunk!
|
|
|
|
There is nothing thinner than 0.3mm in the Openreach network.
I'm also struggling to see how the drop wire to your house is aluminium and is 0.3mm?
Thinnest drop wire in use is 0.5mm copper (CW1378), drop wire 6, 8 etc were even thicker (but not twisted and one of them being copper covered steel).
Drop wire 11 and 15 (used to be called 10 and 14 before the changed 10 to one pair and renamed them both by an extra 1) are 0.5mm copper and have been used more recently.
|
|
|
See the following link for all the old drop wire specifications....
http://www.batt.co.uk/products/view/226/CW1378-Cable...
|
|
|
in use
I think that is the problem in your post. It might be the thinnest that would be used in new or replacement installations, but that doesn't tell you anything about any individual case of the 28 million existing lines.
Presumably the rules were different 50 years ago.
|
|
|
There is nothing thinner than 0.3mm in the Openreach network.
I'm also struggling to see how the drop wire to your house is aluminium and is 0.3mm?
Thinnest drop wire in use is 0.5mm copper (CW1378), drop wire 6, 8 etc were even thicker (but not twisted and one of them being copper covered steel).
Drop wire 11 and 15 (used to be called 10 and 14 before the changed 10 to one pair and renamed them both by an extra 1) are 0.5mm copper and have been used more recently.
BT were deploying .0.32 ali in late 1970s. By coincidence thats when the houses around me were built.
|
|
|
I have to say, I was quite impressed by the results there. I had expected the range of "superfast" to be extended some 200-300 metres, but 1600 metres would be more than I really expected.
And, as you say, interesting implications for BDUK - but equally interesting consequences for the extender technology, like FTTRN. It might alter the planning decisions about where these nodes should go.
There is anecdotal evidence that the rollout seems to have been Huawei-heavy recently. With no visible progress with ECI's vectoring, this might be a reason why.
Yes that's interesting, I've noticed the same. It seems to be mostly Huawei cabs going in at the moment, even in areas where the headend is ECI and all previous cabs have been ECI. It doesn't seem to be a problem mixing cabinet vendors in areas now.
|
|
|
|
The Huawei cabs will be connected to a Huawei headend at the exchange.
|
|
|
|
Not drop wire though......
|
|
|
in use
I think that is the problem in your post. It might be the thinnest that would be used in new or replacement installations, but that doesn't tell you anything about any individual case of the 28 million existing lines.
Presumably the rules were different 50 years ago.
Cables were thicker all those years ago, just not twisted.
http://www.caledonian-cables.co.uk/Telephone/Indoor/...
|