General Discussion
  >> Fibre Broadband


Register (or login) on our website and you will not see this ad.


Pages in this thread: 1 | 2 | 3 | 4 | 5 | (show all)   Print Thread
Standard User Evari
(newbie) Wed 24-Apr-13 23:32:23
Print Post

Vectoring; should I care? (ELI5)


[link to this post]
 
I've been hearing a lot about vectoring on this site recently and apparently it will make speeds of 100mbps possible on the current FTTC rollout. I currently receive ~70mbps on my upto 80mbps BT Infinity line. Would vectoring make any significant difference to me? And if so are BT likely to want me to reset my 18 month contract to take advantage of it?
Standard User ryant704
(member) Thu 25-Apr-13 01:02:42
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Evari] [link to this post]
 
Yes you will more than likely see an increase, not sure about the contract but I would happily say it will take more than 18 months to be rolled out.
Standard User mikehiow
(committed) Thu 25-Apr-13 01:06:47
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Evari] [link to this post]
 
AFAIK, Vectoring isn't even in trial yet (although it's starting this Summer, I think?) so you're not likely to see it for a while.

There's been a lot of speculation about what equipment will need to be changed to enable Vectoring on the current Openreach FTTC kit - new frames/line cards are going to make it less likely/further away. From what I've read, there's a real potential for 120mbps with vectoring but it'd also allow them to offer higher speeds to more people, which may play in their favour in terms of marketing etc.

There should be appreciable gains for most people. When you mention ~70mbps, is that the actual throughput you're receiving? If that's the case, you're likely on the capped sync of 80000kbps with a max attainable already higher than that, which means you'll see a benefit from the higher speed profiles alone, should they ever look at higher speeds with vectoring.

ZeN > plusnet > entanet > <aaisp.net> > Sky LLU > WightWireless > Plusnet FTTC 73/17

Edited by mikehiow (Thu 25-Apr-13 01:08:55)


Register (or login) on our website and you will not see this ad.

Standard User WWWombat
(fountain of knowledge) Thu 25-Apr-13 01:45:58
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Evari] [link to this post]
 
Yes, you probably should care. Even if you don't want to go any faster. Your neighbours want you to care too.

Why should you care? It might not be obvious now, but it probably will become obvious over time.

Back on ADSL, your biggest enemies were the distance from the exchange, and noise caused by line errors or interference.

Now you are on FTTC, and are getting 70Mbps, you've already shown that those 2 factors can be ignored. The replacement factor - that you're close to the cabinet - is less of a factor too. You're probably within 400 metres or so.

But your last remaining enemy, and now the biggest enemy, hasn't shown its face yet... it is the group of other subscribers that are being added to your cabinet over time.

As subscribers are added, the signals to and from their modems are picked up by your telephone line, and appear as noise to your modem. This is called crosstalk - and the more subscribers there are, the more crosstalk you will see. Some subscribers will have little effect on you, while some will have a dramatic effect. Your line is having the reverse effect too.

The added noise from crosstalk means that either you will lose some of your speed, or you will suffer from errors and packet loss. DLM will turn on interleaving & FEC to correct those errors, so you will see increased latency and a reduction in speed.

The effects are random, but mean that any line could lose 40-50% of its theoretical top speed. That might not mean much to people 100 metres from the line (where most people will still get near 80Mbps), but at 1km range, the drop can take people down to 25Mbps.

Vectoring is a smart process to remove the crosstalk, so your modem gets a cleaner signal. There will still be noise from other sources, but the signal should be better. A better signal means fewer errors

Where do things go from here?

Well, with a better signal, you can do two things: a) Run at a higher speed, or b) Run at a lower transmit power.

So, if BT chose to, they could probably add packages at 100Mbps, or possibly even 120Mbps, and let you run at higher speeds. Trials in other countries show that 100Mbps is a plausible speed for properties within 300 metres, and possible within 500 metres, depending on the state of the copper wires. BT haven't run public trials yet, so we don't know how things will fare here.

BT could also choose to not introduce higher top speeds, but to allow modems to run at lower power, at least for closer subscribers. At lower power, it is possible that these subscribers would have even less impact on the more remote customers, especially if vectoring is less than perfect at removing crosstalk. The more remote subscribers would get a better signal, and get a speed boost up to closer to theoretical limit for their line.

For customers between 800 metres and 1km from the cabinet, vectoring could improve their speed from 25Mbps up to 40-50Mbps.

That's important to BT because somewhere in the region of 80% and 90% of the country are within 1km of their cabinets (if they have cabinets).

Finally, the technical problem to all this... Vectoring only works if it can work on all the lines from the cabinet in one go. For BT to get any advantage from this at all, even for a single customer, every subscriber on the cabinet must be vectored.

So, to answer your questions:

- No, it won't make much of a difference to the best performance you are seeing now
- Yes, it could make a difference to you when you cabinet is running full, and your top speeds have fallen away
- It will certainly make a difference to your neighbours living 500 metres further away.
- BT aren't likely to reset the 18 month contract just to put you on vectoring.
- But if BT introduce higher speed packages (say 100Mbps) and you want to be on them, I suspect a move is likely to trigger a new contract.

But... it hasn't even gone to a live trial yet, and I guess that BT will want to try a lot of things out to balance between higher package speeds and lower power; between speed improvements for closer lines and improved range.

The only answer is to keep watching.
Standard User Crusiux
(learned) Thu 25-Apr-13 07:29:48
Print Post

Re: Vectoring; should I care? (ELI5)


[re: WWWombat] [link to this post]
 
When you say that every subscriber has to moved to vectoring on a cabinet for it to work, will they all need a new modem to support vectoring or does the magic take place cab side?

Unlimited BT Infinity 2 Down 73.6 Up 9.7
Standard User Chrysalis
(legend) Thu 25-Apr-13 12:59:21
Print Post

Re: Vectoring; should I care? (ELI5)


[re: mikehiow] [link to this post]
 
according to ispreview BT see it as a speed enabler not speed booster.

Meaning they are not going to increase headline speeds with vectoring but instead use it to allow more lines to hit the full 80/20. Or at least get closer to it. This is a good thing in my view.

Also a ofcom document I read suggests 2013 for start of vectoring rollout. But 2017 as a end year.

BT Infinity 2 Since Dec 2012

Edited by Chrysalis (Thu 25-Apr-13 14:30:15)

Standard User mikehiow
(committed) Thu 25-Apr-13 13:36:48
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Chrysalis] [link to this post]
 
I think it'd be disappointing if BT don't at least boost the speeds slightly.

I'd be surprised if they weren't keen on at least hitting or exceeding 100mbps to match or better Virgin's headline offering.

ZeN > plusnet > entanet > <aaisp.net> > Sky LLU > WightWireless > Plusnet FTTC 73/17
Standard User RobertoS
(sensei) Thu 25-Apr-13 13:44:36
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Crusiux] [link to this post]
 
In reply to a post by Crusiux:
When you say that every subscriber has to moved to vectoring on a cabinet for it to work, will they all need a new modem to support vectoring or does the magic take place cab side?
I assume the OR modems already support vectoring, or would with a firmware update.

The specification for non-OR modems and modem/routers is quite clear.
The modem shall support 17MHz Vectoring as defined in G.993.5[8]. This requires the modem to be “vector ready”.
Source: SIN 498v4p3.

My broadband basic info/help site - www.robertos.me.uk | Domains,website and mail hosting - Tsohost.
Connection - Plusnet UnLim Fibre (FTTC). Sync ~ 54.2/15.2Mbps @ 600m. - BQM

"Where talent is a dwarf, self-esteem is a giant." - Jean-Antoine Petit-Senn.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Allergy information: This post was manufactured in an environment where nuts are present. It may include traces of understatement, litotes and humour.

Edited by RobertoS (Thu 25-Apr-13 14:27:02)

Standard User Chrysalis
(legend) Thu 25-Apr-13 14:34:12
Print Post

Re: Vectoring; should I care? (ELI5)


[re: mikehiow] [link to this post]
 
except they will already beat VMs headline speed with FTTPoD.

BT Infinity 2 Since Dec 2012
Standard User WWWombat
(fountain of knowledge) Thu 25-Apr-13 15:34:09
Print Post

Re: Vectoring; should I care? (ELI5)


[re: Crusiux] [link to this post]
 
In reply to a post by Crusiux:
When you say that every subscriber has to moved to vectoring on a cabinet for it to work, will they all need a new modem to support vectoring or does the magic take place cab side?

The vast majority of the magic happens in the cab for both upstream and downstream.

The user modems are needed to measure the analogue error seen on the line, and report back during initialisation. That helps the cabinet decide how to set up vectoring. This will need at least a firmware update to be considered "vectoring friendly".

It is believed that many modems of today's vintage can cope with this, but I haven't seen that statement expanded into proper details.

Even if they can't, I understand that the cab can cope with less precise (and slower) SNR details that are already seen today. I haven't seen what the real consequences of this are. Details of trials are almost impossible to come by except for those published by the equipment manufacturers, and they (of course) only want to show the positive spin.
Pages in this thread: 1 | 2 | 3 | 4 | 5 | (show all)   Print Thread

Jump to