Kitz Forum
Broadband Related => Broadband Technology => Topic started by: WWWombat on June 10, 2015, 04:04:00 PM
-
My discovery of "profile 35b" (linked here (http://forum.kitz.co.uk/index.php/topic,15577.0.html)) came about when I found some information about the Danish telco TDC, and their vectoring deployment, combined with pair bonding. Some details can be seen in this document, which is in Danish (but Google translate helps in portions):
https://wholesale.tdc.dk/Produkter/Branchemoede_2015_04_28_Produkt_Udviklingsplaner.pdf
Pages 2,13 and 54 show details (including new speed categories up to 150/50) for pair bonding in combination with vectoring.
However, the document also gives details of both their trials with G.fast *and* Vplus/Profile-35b technology.
Interesting... once you can translate it!
The graph shown is a slight variation from Alcatel's original article on Vplus, and now includes the 35MHz performance data...
http://www2.alcatel-lucent.com/techzine/vplus-gets-vdsl2-vectoring/
-
Great image, thanks :D
That makes me wonder just how G.fast works. Serious number crunching hardware at both sides? :)
-
I dont buy the viewpoint that vectoring is not worth rolling out because of g.fast, both technologies have a place.
-
Perhaps it's not worth adding or replacing cabinet hardware to support vectored VDSL2, if instead some of might be replaced by G.fast hardware that could also do vectored VDSL2 for longer lines. It might depend on what the cost of the FTTC cabinet contents was compared to the cost of getting fibre and power to all those cabinets.
-
Interesting graphic.
Out of interest this morning I've been trying to work out exactly why it is I get a rather nice sync (77360/19999) when I'm apparently 450M (straight line) to my cabinet. Obviously it could be slightly more than that - I don't know where the wires go.
Looks like G.Fast will be useless for me though which was what I feared. BUT, I can't complain too much - I do get 77Mbit... Though I'd be interested in 200Mbit vplus......
-
Gfast won't necessarily be useless, all depends where they locate the equipment. I'm also 450m confirmed but currently syncing at 44/6 :-(
-
your sync seems very bad for 450m ronski.
I am close to 400m and have 73/20 with the US having 30 attainable.
Plus part of my line is ali.
Most graphs also rate 450m higher than 44 as well.
-
Gfast won't necessarily be useless, all depends where they locate the equipment. I'm also 450m confirmed but currently syncing at 44/6 :-(
I'm syncing faster and I'm at least 750m away so something seems up with your line.
Have you got any graphs (e.g. Hlog, QLN) to show us?
-
I've had FTTC for almost 3 years now, the sync for the distance has always been bad, I was first on the cab, best I've had is just over 50, worse just under 40. Upstream has halved over the three years, but down stream has generally improved. In the early days I did try and persuade PN/BTor to look into it, but got no where apart from a basic check. It's plenty fast enough for what we use it for, so it's not worth the hassle, I'll just wait for vectoring (eci cab though) or better still gfast.
PS. Stats are on MDSLWB
-
I can't seem to zoom into your HLOG but there appears to be a peak in your US0. Local wiring issue/dodgy extension/master/drop cable?
-
If I get time later and remember I'll start a new thread and post some graphs.
-
Out of interest this morning I've been trying to work out exactly why it is I get a rather nice sync (77360/19999) when I'm apparently 450M (straight line) to my cabinet. Obviously it could be slightly more than that - I don't know where the wires go.
Speed depends on the gauge of the copper as much as the length. Thicker copper can give people better performance than someone with half the length.
The lines that were longest (from the exchange) are probably the ones that stand a chance of involving thicker copper, so if you end up close to a cabinet that itself is far from the exchange, you might benefit.
That isn't the only case though. In a previous property, all the houses on the estate had speed estimates consistent with each other. However, dotted around the development were 5 small blocks of apartments; the estimates for these were considerably better than the nearby houses ... in every single case! The only conclusion I could reach was that they used thicker gauge copper to each block, while all the houses were left on standard 0.5mm copper.
Looks like G.Fast will be useless for me though which was what I feared. BUT, I can't complain too much - I do get 77Mbit... Though I'd be interested in 200Mbit vplus......
As Ronski says, the absolute key thing will be where they choose to put the G.fast nodes. We aren't close to knowing that yet.
-
Just noticed this thread & I used to work on some of this stuff for Alcatel-Lucent (probably now out of NDA as well :D).
Vplus is what I know as "Vectored Phantom" (http://www.uknof.org.uk/uknof21/Maes-Gfast.pdf page 5) and you can continue to bond pairs to get upwards of 800Mbps (aggregate, up to the teleco how they want to split up/down speeds) on multiple 100m loops - I've seen it done. Never going to happen in the UK due to LLU for one thing & BT for another.
Not at all sure where people are getting the idea that g.fast doesn't involve vectoring/pre-compensation - it does as you won't believe the xtalk at 100MHz on old copper assets ;) It also sort of requires that you shut down the FM broadcast band (88-107 MHz IIRC) as well - something multiple Brit govts have been trying to do for years for different reasons. Shurely shome coincidence yes? Actually it is in this instance as all the relevant Brit govts have been interested in is maximising licensing revenue (from the broadcasters) & DAB is seen as the golden goose in that regard (we won't go there, DAB in the UK is awful).
Frankly unless you live in a tower block or in a small village (where houses are close together - inside 200m) I doubt many people in the UK will ever have g.fast available to them this side of 2025.
-
Frankly unless you live in a tower block or in a small village (where houses are close together - inside 200m) I doubt many people in the UK will ever have g.fast available to them this side of 2025.
Due to Talk-Talk - SKY's joint venture in York, I've a feeling G.Fast will be making its presence known waaaay before that estimate, Rizla. ;)
Partial quote below regarding their new Fibre infrastructure.
They have now launched an all-in, entry-level price of £21.70/month for home users and £25/month for businesses, which should make the offer, known as or UltraFibreOptic (UFO), highly competitive
-
I hadn't been paying attention to that but I was under the impression that was FTTP?
-
The York venture is indeed FTTP ....... granted, G.Fast isn't full FTTP but is still classified as 'Ultra Fast Broadband' ..... point being though that this will be rolled out way before 2025, IMHO. ;)
More likely within the next 2yrs or so. :)
-
Totally different then mate.
g.fast is frankly a ludicrous idea for anything other than niche applications & its only interest is to incumbent (monopoly) telecos trying to sweat the last penny from their archaic copper assets.
It WILL end up costing more than simply biting the bullet & installing FTTP but we live in an era of beancounters/obscene director "bonuses" - no doubt the tories will continue to pump taxpayers' money towards BT anyway so BT won't give a toss.
-
it is but BT will be happy to compete with FTTP using g.fast as the consumers will just read the headline speed.
-
I concur with Chrys ............. folk will be more than happy with what G.Fast has to offer IMHO. My main point though was to highlight that it is likely to be rolled out a helluva lot earlier than Riz's 2025 prediction, rather than the technology itself. :) :)
-
I can't see BT installing g.fast nodes every 200m in urban areas - that IS what it'll take to "compete" with FTTP speeds.
g.fast is bugger all use* at distances over 100m unless you start using bonding (and vectoring 2.0) which increases complexity by an order of magnitude. Add in power requirements (reverse power to the remote node won't work in the UK, copper infrastructure is in too poor a state) and it'll cost an absolute fortune just to get on level terms with where Virgin cable speeds are NOW, never mind FTTP speeds.
g.fast has its place but it certainly isn't suited to widespread deployment like VDSL2.
*on UK test loop bundles (ie in the lab) you will get a maximum of 160Mbps aggregate on a 200m line - and that's only if you shut down the FM broadcast band. If BT were capable of deploying vectoring on VDSL2 (I don't believe they are) then you'd get close to those speeds using the 17MHz profile and exceed them using the 30MHz profile.
-
rizla my guess is BT are going to deploy g.fast next year but only from cabinets so very limited coverage.
-
Not my forte at all ...... the technologies that is ........ but our Chief Engineering department, via a nationwide road show for all engineers, were pretty confident G.Fast would work pretty well up to 200/250mtrs and were testing at reaching to 300mtrs using whatever magic these guys have at their disposal.
Forward powering will be the method deployed.
I'm sure I recall something about 106/116Mhz spectrum being involved ?? I may have this wrong, as it is purely from memory and there was a lot to take in.
As we all know by now, VM's network was incredibly selective when first put in the ground. OR don't have the same liberties as such. I don't think the big game plan is to compete at the speed levels VM's network should be capable of, (but isn't, due to poor backhaul), the predictions of 120Mbps DS speed requirements by 2025 is what our boffins mentioned. Having a Gig to hand is pure waste. ;)
However it pans out, the impression I got from them stood there in front of thousands of engineers and all levels of management, was they are fully committed to G.Fast and saying, "I can't see BT installing g.fast nodes every 200m in urban areas - that IS what it'll take to "compete" with FTTP speeds" is of course an opinion, and purely speculative.
I've seen lots of speculation on this forum through debate, most of which tends to be proven otherwise. I hope this is the case with your assumptions too, Riz ??. ;D ;D
-
g.fast has its place but it certainly isn't suited to widespread deployment like VDSL2.
Do you think any has told BT (http://www.btplc.com/News/#/pressreleases/bt-ceo-sets-out-ultrafast-broadband-vision-1116683) though?
-
I (re)designed some of the Alcatel-Lucent g.fast stuff (reduce power/increase vectoring resources on the same die size) so I'm pretty sure I know more than your "Chief Engineering dept" about the technology ;)
At 200m you will get the same speeds with g.fast as you would with 30MHz VDSL2 (vectoring 2.0) at probably 5-10 times the cost in urban areas. End of story.
The ONLY people who consider this to be either sane or cost-effective are incumbent (monopoly) telecos desperately clinging onto their copper assets.
There is a reason Alcatel-Lucent are up for sale (at any price - Bell Labs not included AFAIK) and its because no matter how clever they are & how complex the dies become, their business is dying. FTTP offers far greater speeds with far less complexity.
g.fast has its place (there are a lot of villages it'd work very well in) but its insanity on wheels to present it as the "future". It isn't & never will be.
-
g.fast has its place but it certainly isn't suited to widespread deployment like VDSL2.
Do you think any has told BT (http://www.btplc.com/News/#/pressreleases/bt-ceo-sets-out-ultrafast-broadband-vision-1116683) though?
That PR blurb is pure comedy gold : "The UK is ahead of its major European neighbours when it comes to broadband" :D
So we're ahead of (for example) Telecom Eireann? Like hell we are - they managed g.inp AND have rolled out vectoring across the network. Aggregate speeds of 140Mbps on lines of 100m are not unusual. Even ALBANIA managed to rollout g.inp and vectoring.
BT are delusional.
-
not ahead of the Netherlands either.
-
g.fast has its place but it certainly isn't suited to widespread deployment like VDSL2.
Do you think any has told BT (http://www.btplc.com/News/#/pressreleases/bt-ceo-sets-out-ultrafast-broadband-vision-1116683) though?
That PR blurb is pure comedy gold : "The UK is ahead of its major European neighbours when it comes to broadband" :D
So we're ahead of (for example) Telecom Eireann? Like hell we are - they managed g.inp AND have rolled out vectoring across the network. Aggregate speeds of 140Mbps on lines of 100m are not unusual. Even ALBANIA managed to rollout g.inp and vectoring.
BT are delusional.
I wouldn't class Eire or Albania as 'Major'. Different rules for itsy-bitsy size-of-a-stamp countries. Is it Guernsey/Jersey (?) that some folk like to use as a comparison to the UK with its FTTP infrastructure ?? Same with this instance, Rizla. :)
-
not ahead of the Netherlands either.
Whoops ............... forgot those giants. ;) ;D
-
Or Austria. Or France. Or Spain (yes really). Or Sweden, Norway, Iceland, Finland. Admittedly we are "ahead" of Italy & (for a few months) Germany. Edit - to be complete we are also "ahead" of Slovakia, Greece & Bulgaria (not by much in Bulgaria though).
That's OK though, you carry on with your "little Englander" viewpoint....
-
The way I look at it is that FTTC got fibre out into the streets, and g.fast will move it further out and far more wide spread, each step getting closer to the end user. Look at it as another step in the direction of full FTTP
Yes it may well cost more in the long run, but at least they have income coming in from each step, which offsets it.
-
Or Austria. Or France. Or Spain (yes really). Or Sweden, Norway, Iceland, Finland. Admittedly we are "ahead" of Italy & (for a few months) Germany. Edit - to be complete we are also "ahead" of Slovakia, Greece & Bulgaria (not by much in Bulgaria though).
That's OK though, you carry on with your "little Englander" viewpoint....
Oooh touchy. I'll concede at this point as you point-blank refuse to accept what BT are saying or have planned for the future. You obviously have access to more info regarding statistics than the relatively unknown, small outfit known as BT Group.
I also hope our Chief Engineering Department (i.e.: Ted and Jack - a couple of retired bus-drivers who potter around in a shed at the bottom of Ted's garden), are looking in and seek advice if they need it, regarding G.fast.
As you were.
-
The way I look at it is that FTTC got fibre out into the streets, and g.fast will move it further out and far more wide spread, each step getting closer to the end user. Look at it as another step in the direction of full FTTP
Yes it may well cost more in the long run, but at least they have income coming in from each step, which offsets it.
Agree Ronski, and a rather more realistic view-point of what is happening in the comms world. Whatever it does cost, BT have a commitment first and foremost to their shareholders …… not a few contributors on a few forums. :)
-
Both ronski and rizla got valid points.
I agree with ronski that g.fast (if BT actually extend the fibre, I think they will just deploy from cabinets) will be a step closer to FTTP.
But, Rizla's point has a lot of merits and I am sitting here dumbfounded as to why BT havent bothered with vectoring, its a logical step forward which shouldnt have a large cost. BT proved somewhat inept when they they did their limited rollout of g.inp which should have been a fairly trivial deployment. What I have observed with BT is instead of doing regular incremental improvements to their technology they instead prefer to have large gaps between improvements and have a large step in improvement when they make those improvements, plus the obsession with modifying technologies that work well elsewhere to their own variant does them no favours.
-
Where does it say BT are 'not bothering with vectoring' ?? This is a complete un-truth. Either that, or they have just sent me a Broadcom SIM (at a cost of nearly £1,000) to use on Vectored Cabs, as their way of having a joke. Times that by 'x' thousand engineers requiring the upgrade ...... and they're having a real laugh.
Or, maybe vectoring is being deployed ?
-
Ok "if" BT dont deploy vectoring, but as it stands they have yet to deploy it. I can only speak for what is happening now and their public announcements. :)
-
What are they doing in the G.Fast trial areas, just cabinets or extending further out?
The map on this page (http://www.ispreview.co.uk/index.php/2015/04/bt-confirms-third-500mbps-uk-g-fast-broadband-pilot-for-swansea.html) shows that they are extending out with the DPU's being G.Fast Distribution Point Units.
-
Where does it say BT are 'not bothering with vectoring' ?? This is a complete un-truth. Either that, or they have just sent me a Broadcom SIM (at a cost of nearly £1,000) to use on Vectored Cabs, as their way of having a joke. Times that by 'x' thousand engineers requiring the upgrade ...... and they're having a real laugh.
Or, maybe vectoring is being deployed ?
So,reading between the lines then Black Sheep, vectoring must be close (very close :hmm:) to being deployed. ;) :o
-
What I have observed with BT is instead of doing regular incremental improvements to their technology they instead prefer to have large gaps between improvements and have a large step in improvement when they make those improvements, plus the obsession with modifying technologies that work well elsewhere to their own variant does them no favours.
Where is BT doing this? I thought the fact they were waiting until standards were set/kit available was one of the factors slowing down new things - and other (non UK) telco's tend to be more willing to deploy based on draft standards.
-
What are they doing in the G.Fast trial areas, just cabinets or extending further out?
The map on this page (http://www.ispreview.co.uk/index.php/2015/04/bt-confirms-third-500mbps-uk-g-fast-broadband-pilot-for-swansea.html) shows that they are extending out with the DPU's being G.Fast Distribution Point Units.
thanks ronski, first time I seen that image.
-
Fibre-to-the-Remote-Node (FTTrN) as it's known, where does the word G.Fast come in ?
-
Not sure I know what you mean NS?
-
Interesting debate since I was last here!
Riz - I get the impression your work with A-L is not necessarily current. Is that true? Or are you relating information about state-of-the-art?
The impression I got, from reading BT bumph, is that that they're aiming for a self-styled "multi-100-megabit" at first, which I read to mean 200-300Mbps. The stuff on 500 Mbps seems to be toward the end of their 10-year vision. My take on their change of emphasis is that they now think a rollout based (initially) on existing cabinets is a worthwhile strategy ... suggesting that range is perceived to be better than expected. The higher speeds can be left to the second half, with deeper nodes.
Relating to this, both BT and Sckipio seem to report that G.Fast is behaving better than they expected. Sckipio seem to report good results (though I'd expect a manufacturer to not be immune to hype). The earlier reports are encapsulated in this article:
http://gfastnews.com/index.php/90-r/157-suddenly-g-fast-is-400-500-meters-not-100-200-meters
Since then, Sckipio have made a presentation at one of this year's major G.Fast conflab's. In the presentation audio (from youtube, towards the end), you can hear the references to where G.fast can be taken in the future ... and, in particular, hear the references to BT targeting longer range.
https://www.youtube.com/watch?v=HPDRS06t5Kg
What I take from both of those happens to match what I see from BT. As an example, see this presentation from UKNOF:
https://www.youtube.com/watch?v=LLpk2dz6nBQ
Here you can see that their 500Mbps aims are related for G.fast mk.II, and not the current version. But the strategy and direction seem clear.
(@Chrys - that map you hadn't seen before comes from this presentation)
The most recent article I've seen is this one:
http://gfastnews.com/index.php/90-r/170-more-megahertz-g-fast-s-simple-story
I see promising things appearing in a plausible pipeline. But I don't doubt that the economics hinge around some very tight criteria, and that some things won't pan out as well as hoped.
@NS - FTTRN is BT's terminology for a small VDSL2 cabinet sited deeper in the network, to suit premises that are more than 1km from the existing FTTC cabinet. It still aims at today's superfast packages, up to 80Mbps, and has nothing whatsoever to do with G.fast or FTTdp.
-
Wow ...... great post ..... very informative. :)
-
so cabinet only in early rollout, then the extra nodes latter part of rollout?
-
That's my interpretation, based on reading a variety of BT statements alongside a number of analysis pieces. It might not be quite so black and white in the end, though.
And, IMHO, the biggest issue isn't about the telco side of things, but about humble power.
My understanding of the current FTTC rollout (in both the commercial and BDUK portions) is that power is one of the biggest problem areas. Even as standard (a one-of payment of around £1,000), the cost of power installation is one line item that makes or breaks a cabinet; excess costs can easily bring the cabinet out of the project - and this can happen if the power company passes on costs, eg for new transformers. As an example, 3 cabinets were found in one town in my county to need £90,000 to get power to them; they were dropped from the project until the LA used some "leverage" to persuade BT to include them.
The power problem extends to FTTRN, and is what has stalled progress there. The only existing/known rural node needed a separate power cabinet building for it.
If we assume that a G.Fast rollout will be plagued by the same issue, I think we are standing on safe ground.
So, having said that, it is easy to see that BT can probably manage to deploy a G.fast node at the existing FTTC site without worrying about the cost of power, as they can feed it off their existing one. However, any extra nodes will either need local power of their own, or will need "forward power" fed over spare copper from the current node, or will depend on "reverse power" fed from the end users.
Deploying extra nodes will become a trade-off between the distance away from end-users (fewer nodes, slower speeds) vs the distance away from the existing cabinet ("forward power" losses depending on length of copper) vs the maturity and acceptability of "reverse power" solutions.
Cabinet-only becomes a relatively easy no-brainer, while extra nodes need some finesse.
But there's an upper limit, financially, to all this: there's no point spending more (in any one area) than an FTTH rollout would cost.
-
This is a shot in the dark, but can anyone find what the maximum tone number and/or the tone width figures are for 35b? I assume its 35Mhz top frequency.
-
I don't know the tone numbering, but I think you're right that it is in the region of 35MHz.
However, one of the major points that Alcatel-Lucent were making in their push of Vplus (at 30MHz) was that it was "vectoring-compatible" with profile 17a. That compatibility required the tone width to be identical between the two profile; profile 30a is vectoring-incompatible with 17a because it has a different width.
I suspect that 35b will have identical channel width to 17a: 4.3125 kHz
Then, because 17a has an upper limit of 4096 tones, it actually ends at 17.664MHz. It really wouldn't surprise me if we discovered that 35b actually used 8192 tones, and stops at 35.328MHz
-
Given ignition now seems to be on a vectoring trial, dont know why he just gives tidbits of information but he revealed yesterday he is on it.
I think openreach are still considering a commercial rollout of that tech, in my mind if they were only going to roll it out to bduk areas then they would be testing it in those areas only.
-
I think some were already aware, but here's a snippet of a latest brief ....
Trials of G.fast will start on 17 August 2015 and will be held in Huntingdon, Cambridgeshire and Gosforth, Newcastle (in certain postcode areas); whilst fibre on demand trials will take place in Huntingdon only.
-
Given ignition now seems to be on a vectoring trial, dont know why he just gives tidbits of information but he revealed yesterday he is on it.
I must have missed this entirely. I can't find *where* though.
If he is on a trial, perhaps it is for node-scale vectoring. I thought his cab had outgrown itself, and even had an extension on the PCP.
Trials of G.fast will start on 17 August 2015 and will be held in Huntingdon, Cambridgeshire and Gosforth, Newcastle (in certain postcode areas); whilst fibre on demand trials will take place in Huntingdon only.
Not long to go now.
-
My area is one of those in which the only real solution is FTTP or FTTRN.
One cabinet is at least 750m away from most residents (I'm one of them): it serves premises a good 1.5KM away.
The other cabinet is at least 1km away from most residents; it serves residents a good 2KM away. FTTC isn't going to help on that second cabinet as most people will be too far away from it to benefit.
-
The way I look at it is that FTTC got fibre out into the streets, and g.fast will move it further out and far more wide spread, each step getting closer to the end user. Look at it as another step in the direction of full FTTP
Yes it may well cost more in the long run, but at least they have income coming in from each step, which offsets it.
G.fast at my cab will suit me fine i'm within 200m. However anything that may help bring a real fiber optic connections closer to most houses cannot be a bad thing. Especially with possibility the G.fast modules being setup in a way that allows for Fibe on demand installations. That was surely BT's dream, be be able to build a network that makes Virgin look old hat.
I've also gotta say at least TDC are not just thinking inside the traditional box with regards to DSL. Then again unlike BT, most other providers don't struggle with technologies like G.INP or vectoring.
-
Given ignition now seems to be on a vectoring trial, dont know why he just gives tidbits of information but he revealed yesterday he is on it.
I must have missed this entirely. I can't find *where* though.
If he is on a trial, perhaps it is for node-scale vectoring. I thought his cab had outgrown itself, and even had an extension on the PCP.
Trials of G.fast will start on 17 August 2015 and will be held in Huntingdon, Cambridgeshire and Gosforth, Newcastle (in certain postcode areas); whilst fibre on demand trials will take place in Huntingdon only.
Not long to go now.
???
Are the trials go so well, that they are bringing forward the release then. :hmm: :hmm:
-
No idea ........... a trial tends to last far more than just 1 month, and with the G.fast trial only starting on the 17th August 20115 (as highlighted by yourself in the above post) ........... then I wouldn't expect answers anytime in the very near future.
Expectations are for G.Fast to reach over 10 million EU's by the end of 2020 though.
-
the G.fast trial only starting on the 17th August 20115
We have a hell of a long wait then :P
-
the G.fast trial only starting on the 17th August 20115
We have a hell of a long wait then :P
Ha ha ...... it took a minute or two then for the penny to drop. ;D ;D
-
Bit of further info .......... trial dates are, 17th August 2015 in Huntingdon and 21st September 2015 in Gosforth. The trials are expected to be approximately 6 months in duration.
Hopefully there will be a small scale roll-out within 2016.
-
not that long a wait given openreach have already said they starting g.fast rollout in 2016. g.fast doesnt excite me tho so i will be keeping my vdsl service. Especially as g.fast will have less favourable contention.
-
And more good news BT is rolling IPv6 to Entire Network by December 2016.
http://www.ispreview.co.uk/index.php/2015/09/uk-isp-bt-to-deploy-ipv6-to-entire-network-by-december-2016.html
-
not that long a wait given openreach have already said they starting g.fast rollout in 2016. g.fast doesnt excite me tho so i will be keeping my vdsl service.
My understanding is that G.Fast will be implemented by bringing the fibre nearer to the EUs with the use of mini/micro-DSLAMs, i.e. FTTN (where N is node). VDSL2 will still be used over the last metallic segment in the pathway.
Or am I totally confused? ???
Especially as g.fast will have less favourable contention.
Do you have a link to a reference explaining the above, please Chrys?
-
And more good news BT is rolling IPv6 to Entire Network by December 2016.
http://www.ispreview.co.uk/index.php/2015/09/uk-isp-bt-to-deploy-ipv6-to-entire-network-by-december-2016.html
Thank you for that link. Perhaps we should be cautious and insert the word "aims" into that headline . . .
UK ISP BT Aims to Deploy IPv6 to Entire Network by December 2016
Many a good plan has suffered subsequent delays due to unforeseen circumstance. ;)
-
not that long a wait given openreach have already said they starting g.fast rollout in 2016. g.fast doesnt excite me tho so i will be keeping my vdsl service.
My understanding is that G.Fast will be implemented by bringing the fibre nearer to the EUs with the use of mini/micro-DSLAMs, i.e. FTTN (where N is node). VDSL2 will still be used over the last metallic segment in the pathway.
Or am I totally confused? ???
Especially as g.fast will have less favourable contention.
Do you have a link to a reference explaining the above, please Chrys?
That's absolutely correct, Cat. There will be no adverse effect on existing VDSL1 services (ie: crosstalk), as there will be frequency separation between the end of VDSL1 tones and the start of VDSL2 tones. :)
-
I just have hearsay from others on tbb.
So we have currently 40/80mbit users on a 2.5gbit cabinet connection.
G.fast it was said would be 300-500mbit on a 1gbit connection from the node. So in theory 3 users could saturate the node, of course the node sizes will be less users than cabinets, but that kind of contention is what cable uses and is too much for me.
Blacksheep openreach dont use vdsl1, only vdsl2, so I assume you mean they wont use the vdsl2 range on g.fast.
-
Apologies. yes, that is indeed what I meant.
-
no worries
-
Frankly I'd be more concerned about what frequencies they intend using as the VHF (87.5-108MHz) band is out, so is the first 17MHz due to VDSL2 and with Ofcom wanting to use the bands 58-68MHz, 70.5-71.5MHz & 80-81.5MHz for the "Internet of Things" <spit> I'm intrigued how BT propose to actually achieve the aggregate speeds they appear to be promising on g.fast.
I suspect that the headline aggregate speeds will be extremely dependent on location - and by that I mean what TXs are radiating in the local area rather than the distance to the node......
-
I just have hearsay from others on tbb.
It is always sensible to keep a proverbial pinch of salt to hand (or paw) when considering such hearsay from that site. ;)
-
Frankly I'd be more concerned about what frequencies they intend using as the VHF (87.5-108MHz) band is out, so is the first 17MHz due to VDSL2 and with Ofcom wanting to use the bands 58-68MHz, 70.5-71.5MHz & 80-81.5MHz for the "Internet of Things" <spit> I'm intrigued how BT propose to actually achieve the aggregate speeds they appear to be promising on g.fast.
I suspect that the headline aggregate speeds will be extremely dependent on location - and by that I mean what TXs are radiating in the local area rather than the distance to the node......
Not my forte, but 20-106Mhz will be used during the trial period. What reality brings I wouldn't know ??
-
Not my forte, but 20-106Mhz will be used during the trial period. What reality brings I wouldn't know ??
That is interesting. I shall speculate that possibly the 87.5 - 106 MHz segment will be masked, with a power cut-back in place. :-\
-
I wouldn't have the foggiest, my learned companion. ???. :)
-
Well, 87.5 - 108 MHz is the range of frequencies allocated to Band II (audio) broadcasting. :)
If the infrastructure will be such that no RF energy escapes from the metallic pair and, thus, does not cause interference to licensed users of the above frequency range all will be good. ;)
-
You're looking at it the wrong way around ;)
Nearby FM broadcasts will play havoc with bitloading even on brand-new cabling with perfect balance - been there got that T-shirt in 2013.
What you need to remember is that the front end on xDSL is differential and if anyone knows how to perfectly balance the diff amps on every die such that no breakthrough occurs then speak up because I can get you a £300k/year job right now with Bell Labs. No joke.
-
Bow out time ..... you two are speaking on another level. ;D ;D
-
You're looking at it the wrong way around ;)
Pah! :D
I can only see xDSL making use of transceivers at each end of a metallic pathway. As long as signals from those transceivers do not leak from the metallic pathway and interfere with licensed broadcast transmissions, all well and good. :)
For those on-lookers, of course I know and understand the points Rizla has made (and I strongly suspect that the reciprocal is also true). Looking at it from my viewpoint, the sooner full fibre optical links are used, the better for all concerned. ;)
-
The problems with g.fast are largely on the analogue side of things but these problems have a HUGE impact on the vectoring resources (DSP side of things) which in turn has a huge impact on die size and power consumption hence the industry interest in analogue engineers.
The bottom line is that all this complexity/cost/power usage is because we're trying to send a signal down a transmission line which it was never designed for. Internal wiring is going to be a nightmare for most people with g.fast IMHO as the vast majority of it will need replacing with cat5 (or equivalent) cabling.
BT want to sweat the last penny out of their last-mile monopoly infrastructure & I'm sure we can all see the logic in that for BT. Remember that companies in the UK have a legal duty to act solely in the interests of shareholders. However for the rest of us then its complete lunacy on wheels :)
As I keep saying g.fast has its place but widescale deployment is NOT that place.
We'll see what we see I guess...
-
We'll see what we see I guess...
And that is a statement in which I concur. :drink:
-
You're looking at it the wrong way around ;)
Pah! :D
I can only see xDSL making use of transceivers at each end of a metallic pathway. As long as signals from those transceivers do not leak from the metallic pathway and interfere with licensed broadcast transmissions, all well and good. :)
Indeed & I rather suspect it will be an issue in terms of interference in (for example) terraced houses where the master socket can be some distance from the entry point of the drop cable. The cabling there is likely to require replacing or it'll cause some issues for neighbours listening to FM.
If the cost of remedial work (for g.fast to get to the headline speeds) facing the consumer was more widely appreciated then I think people would stop wittering on about how expensive proper fibre deployment would be.
As an aside a friend in Sweden has had some disruption to his life as the owners of the block of flats ripped out & replaced "old" cabling. The building has had FTTB since 2009 or so with the service being something like 100/40Mbps max (no problem for anyone). They replaced the VDSL2 box with a super-duper g.fast box and the speeds were "somewhat underwhelming" according to my friend - I think he got an extra 20Mbps aggregate, can't remember but it wasn't much. There was a fair bit of sync instability as well for some flats. Anyway he sent me a picture of what had been installed when he was at work & its got a fibre connection now as well - owners covering their options this time. The block of flats is 11 years old. Edit - he now has a 200/100Mbps service (he can choose how to split the 300Mbps aggregation he pays for). I think in the couple of decades I've known him I have got to within 20Mbps of his downstream once :D Edit2 - the thing that really irritates me is that his brother lives way the hell up in the north of Sweden and still has faster broadband than my 80/20....
-
A clue as to how BT may be confident of getting those speeds even with all the masking and interference is they pushing to change the spec of g.fast to allow higher bitloading.
Also another clue is they pushing for bigger nodes, more user's per node which worryingly probably also means longer copper length per node.
BT could approach this either by just deploying from cabinets or nodes only near cabinets for minimum cost, or they may do what people who are thinking positively think they will do which is push nodes right out within 100m of each property, but if they do the latter the cost wont be much cheaper than FTTP which is why I think they wont do that.
-
Some info here on BT G.Fast and the new version of FTTPod
http://blog.thinkbroadband.com/2015/09/g-fast-and-fod2-a-peek-at-the-openreach-hardware/
-
> Looking at it from my viewpoint, the sooner full fibre optical links are used, the better for all concerned. ;)
Hallelujah!
-
Vplus is being released in my area, will the HG612 support it? Does it just need the G.INP firmware?
http://manx.net/isle-of-man-news/83167/manx-telecom-invests-in-faster-fibre-broadband
-
Vplus is being released in my area, will the HG612 support it?
I very much doubt it. The HG612 is just a modified HG610 which, in turn, is really quite elderly.
Does it just need the G.INP firmware?
No, I don't think any firmware update will confer extra abilities to that old hardware.
-
The HG612 can probably handle enough, on 17MHz, to reach 100Mbps.
However, IIRC, the RJ45 connectors are 100Mbps, not gigabit. You couldn't access the 200Mbps package even if the modem hardware could cope with that sync speed.
-
Thanks for the response, I'll be quite happy with 100 assuming I can get that, I am definitely not paying a further premium for 200. I was syncing at over 80 when I first got VDSL dropped over time to 60.
At 300 - 400 Metres from the cabinet hopefully vplus will at least bring me 80 back.
-
Vplus uses profile 35b which the HG612 wont support given its age
-
Indeed, the HG612 can support profile 30a, but not the newer profile 35b (VPlus).
It just won't sync at all. You will need to buy a much newer modem that lists VDSL2 profile 35b/VPlus support.
-
My understanding is I will still benefit from the vectoring even if I cannot use 35b?
What similar device would do the job? My vdsl plate is in my garage with the modem on the wall above it, I have my main router in the centre of my house.
I have an old n66u doing nothing
-
It sounds to me like they are just uncapping the 80mb service to 100mb. They may be introducing vectoring to help achieve this. Your existing VDSL2 hardware should be able to use this.
The new 200mb service will be the VPlus/profile 35b. This will require new hardware, but the ISP is likely to supply it.
The above is just an educated guess. The article you linked to doesn't give much technical detail.
-
I'll keep an eye out for more tech details and post when available, I'll also grab my stats now as not done for ages, and again when changes are implemented and post the differences.
The ISP's here deploy Fritzboxes now in the main, but the main ISP originally supplied the Netgear DGN3700 for VDSL (With the original Netgear firmware which did support vdsl before being removed in a later update) which I would guess a fair few people still have so don't know how that fits in. I still have one at work, will see if I get offered an upgrade.