G.fast on DSL Checker

For anyone that may not have noticed, G.fast has been added to the checker for those that will be able to get it but where the cabinet is not yet live (planned vs available).

This appears to be a new addition over the last week.

Edit: It seems not everyone is on there but certainly more lines are reporting as planned than before.

Matt

Edited by uno (Mon 12-Mar-18 22:05:28)

I've been checking my new address (well, phone number) every so often but I don't think a pod is even planned here yet as I've tried an address check for houses very close to the cabinet and g.fast isn't listed. I'm hoping at around 350 meters, I believe anyway, I might be within reach of it when it does become available but I have some doubts.

Not the case here unfortunately, we've had a pod since July last year and the Openreach work was done in the 1st week of August (i spoke to the engineers when they were doing it), but there is no listing on the checker for G.fast

Nothing on checker here.

A cabinet near me, which had a pod installed at the end of last year, now has G.fast with availability 'planned' in the checker (it wasn't showing last week).

However a check by address, and the odd known number, around the area seems to show that unless you have 80 for the High VDSL Range A you don't have a chance of getting G.fast, and even then not for certain.

These two properties are next door to each other (despite the numbering):
#37
VDSL A: 80-60.4 / 20-19
VDSL B: 76.9-50 / 20-16.8
G.fast A: 330-330 / 50-34.5
G.fast B: 318.9- 249.6 / 46.1-23.7
#45 (by address)
VDSL A: 78.5-55.4 / 20-18.7
VDSL B: 69-42.1 / 20-12.8
No G.fast
#45 (by phone number)
VDSL A: 80-58.5 / 20-19
VDSL B: 73.6-47 / 20-15.1
No G.fast

These are in the a building of nine flats on the other side of the road:
#1
VDSL A: 80-69 / 20-19
VDSL B: 80-63.2 / 20-19
G.fast A: 330-330 / 50-36
G.fast B: 328.7-260.5 / 47.6-24.8
#9
VDSL A: 80-66.9 / 20-19
VDSL B: 80-59.8 / 20-18.3
G.fast A: 228.6-200.3 / 32.8-14.1
G.fast B: 180.3-124.7 / 22.2-10.7
and the next property in the direction away from the cabinet has no G.fast.

I did some research using the address checker, and it seems that G.Fast drops off massively in speed estimates for neighboring properties, and is only available around 100m for the cab as the crow flies. Might be the wiring from the cab I was looking at though.

The technology seems to me to be a huge waste of money by Openreach.

Or a cheap way of boosting speeds in the short term for areas while the slower to build FTTP network work is underway in other areas.

If Openreach really want to waste money, can I suggest they commission a thinkbroadband study into the link between speeds available to people and what they actually buy, a price of £120,000 would make me happy and make up for shortfall in salary over last six years

For the cab I was looking at the cutoff point appears to be around the 200m mark.

Something has changed with the checker recently in general as I dont have any FTTC/P, I actually show Exploring Solutions.
But now it rubs salt into the wound by telling me my FTTC estimates if Openreach could be bothered to install a cab, instead they cancelled the upgrade from last year and waited for a community fibre scheme to hopefully get them £109k instead.

g.fast is a huge waste of time and money in my opinion, Virgin already are hitting 390Mb with their VIVID350

and BT are trialing a technology that 95% wont even get near that from.

The money spent on FTTC/G.Fast should of just went for a FTTP option from day one, but oh well.

Hi

I completely agree in that G.Fast is a huge waste of money. The only people that will benefit are those that are already on FTTC 80/20, which is plenty fast enough for now and in the short term.

It does nothing to extend the reach of fibre closer to peoples homes, and does nothing to improve speeds for those who can't get anything fast now. (As it seems the idea of small G.Fast nodes fed by fibre moved closer to homes is largely or completely out of the picture now.)

The investment should have just gone into FTTP, and if they wanted to boost speeds of FTTC in the meantime then those close enough to get a speed boost with G.Fast, could benefit from a VDSLMax type of sync where their modem is allowed to sync upto the maximum obtainable (for some reaching 100Mbit/sec) and sorting out vectoring so that everyone else got a boost as well.

We all end up paying for the folly that is G.Fast in higher prices.

Regards

Phil

I largely agree. As originally touted it made sense - roll it out in rural not-spots and slow-spots where a small pod could be installed more cheaply than a large cabinet. Unfortunately I suspect the distance drop-off meant it wouldn't be suitable for that. But that same distance drop-off also doesn't do urban areas any favours.

So, yes, enough already. Give up and spend the resources on FTTP.

In reply to a post by Ixel:

I've been checking my new address (well, phone number) every so often but I don't think a pod is even planned here yet as I've tried an address check for houses very close to the cabinet and g.fast isn't listed. I'm hoping at around 350 meters, I believe anyway, I might be within reach of it when it does become available but I have some doubts.

The impression I have is that anyone currently syncing in the 70s or higher ought to see some benefit from G.FAST. If it plays well with VDSL and assuming that the vectoring proves effective it may be that anyone currently seeing 60 or higher will gain some benefit.

But overall I'm still of the opinion that it's a dead-end technology with insufficient practical advantages. Should probably have been dropped during the research phase and the resources redirected into FTTP.

Edited by Andrue (Wed 14-Mar-18 10:34:06)

I'm synching at 80/20 (albeit at 3db) and I can't get it.

Apparently my line length estimate is ~500m, for a 300m walk.

In reply to a post by Andrue:

In reply to a post by Ixel:

I've been checking my new address (well, phone number) every so often but I don't think a pod is even planned here yet as I've tried an address check for houses very close to the cabinet and g.fast isn't listed. I'm hoping at around 350 meters, I believe anyway, I might be within reach of it when it does become available but I have some doubts.

The impression I have is that anyone currently syncing in the 70s or higher ought to see some benefit from G.FAST. If it plays well with VDSL and assuming that the vectoring proves effective it may be that anyone currently seeing 60 or higher will gain some benefit.

But overall I'm still of the opinion that it's a dead-end technology with insufficient practical advantages. Should probably have been dropped during the research phase and the resources redirected into FTTP.

Hi

In reply to a post by AndyPandy:

I'm synching at 80/20 (albeit at 3db) and I can't get it.

Apparently my line length estimate is ~500m, for a 300m walk.

I'm in a similar situation where I get 80/20 with a bit to spare but being at around ~350 metres may not be offered G.Fast. Of course distance isn't the whole story, as line attenuation (conductor size/material) and cross-talk or lack of play into as well. I guess it depends if they decide based on line length alone.

The other issue is, as I understand it, that it's G.Fast - VDSL, by that I mean due to issues with cross-talk with VDSL, that G.Fast starts from frequencies above VDSL, so G.Fast has a handicap straight away compared to VDSL and has lost 80Mbit/sec of potential speed immediately even if the line is a metre long. This is why VDSL will be faster than G.Fast at longer line lengths and has a much longer reach, as it uses futher travelling lower frequencies that G.Fast can't use due to affecting VDSL. A problem largely avoided if G.Fast was deployed as it was designed to be, much nearer to the premises.

I think they are looking into G.Fast sharing these lower frequencies but I'm not sure it's happening yet if at all. Also experience tells us if a technology isn't in the pilot and first stages of a commercial roll out by ShortReach (sorry OpenReach), the chances of it being added later are remote, as can be seen from the lack of vectoring on VDSL and G.INP missing from ECI cabinets.

So to sum up, we've got a dirty hack that enables data to be sent over decades old telephone cable only designed for telephone calls, that leaks most of that data into fresh air or someone else's line before it reaches our homes, and now we have G.Fast, that has been changed and hacked around with to make it Long Range G.Fast, over the same old cable which benefits only those that don't need the benefit.

I guess they call that progress

Regards

Phil

If you can't get it at those sync speeds then [censored] is the point of it? I'd assumed it might at least uplift 50% of lines on a cabinet but it's now sounding like a lot less than that. Openreach presumably know what they are doing but it sounds like pointless madness. I always said it was a poor idea and now it looks like a bad one. I wonder if perhaps someone at Openreach has too much invested in G.FAST and is ploughing ahead to try and avoid looking stupid.

Hi

I wonder if perhaps someone at Openreach has too much invested in G.FAST and is ploughing ahead to try and avoid looking stupid.

I think there is an element of that. Originally G.Fast was in R&D as nodes closer to homes and extending fibre outwards, which ultimately is an investment that pays back later with full FTTP deployment. The hardware for G.Fast was small nodes supporting a maximum of a dozen or so connections. The problem was the cost in deployment and accountants/investors saying no, they then also had competition from the likes of Virgin that can advertise and sell much faster speeds, so they needed to quickly get a product out there with >100Mbit/sec speeds but as cheaply and quickly as possible. It didn't matter that its reach was short and it didn't speed up slower connections.

OpenReach went back to the equipment suppliers with their new requirements and the equipment manufacturers went back to the drawing board to create Long Reach G.Fast, as the problem was at that point no hardware could support the higher number of connection points required if moving the equipment all the way back to the cabinet, as that was never in the design. There are still issues with vectoring and the amount of processing power required to deal with vectoring over 32/64 or 96 ports which OpenReach were demanding the hardware needed to support.

The irony is VDSL could have delivered similar speeds (although still requiring cabinet replacements), by using the same higher frequencies, but it does seem doesn't it that they had spent so much money on R&D on G.Fast it became a runaway train that couldn't be stopped without a lot of embarrassment and questions asked, so they would hack it into our old corroded copper wires no matter what.

It just means the fastest connections get faster for those people that probably don't need the speed increase (and may not even bother upgrading by and large), and the money and man hours are diverted away from FTTP. Investment in G.Fast is a complete waste of money, at least VDSL brought fibre out of the exchange and closer to homes, but G.Fast does none of that.

Regards

Phil

Edited by deleted (Thu 15-Mar-18 08:28:14)

I along with many others of you have been checking dslchecker daily for about 4 months now. Until Monday night there was no reference to g.fast against my number. Then on Monday I got a 'planned' availability date with a clean range of 330/330 50/35.2 and impacted of 305.8/234.2 44.5/23. However, since Monday night, these results have dissapeared from dslchecker - I've been checking daily again...

I dont know what that means, whether the planned status have been removed from my number, whether it will come back as available, etc. For what its worth I get 79/19 fttc sync rate and am about 180m (walking route) from the cabinet, so sureley a prime g.fast candidate?

Hi

Could just have been someone making a mistake updating the database, probably did a blanket update to show the obtainable speeds which they can work out for all cabs, without actually restricting the updated to only those cabinets that will get a connection. I work in IT and these sorts of mistakes happen all the time.

Regards

Phil

I would be really wary of the checker results as they do not appear to be based on the physical world.

Example. There are over 26 wires/cables from our pole. On our side of the road from No 28 to number 54 with all 14 houses all overhead.

Those physically closer to the cab but further from the pole have a higher speed shown than those closer to the pole ( Shorter line distance).

Clean speeds
28 :- 281-250
30:- 286-253
32:- 316-282
34 shows no G-fast ( may not be OR served, but wires exist)
36/38/40:- 214-187 (closest to the pole)
42:- 221-194
46- 214-187
44 shows no G-fast ( may not be OR served, but wires exist)
46- 214-187
48 up show no g-fast but I expect these are to far from the cab.

IF based on the real world I would have expected those further from the pole to have lower speeds than those close to the pole.

Yep. Showing as planned here

https://ibb.co/i2UuqH

How many cables go down the pole from the distribution blocks on the top of the pole? If it�s serving more than 20 customers it�s likely to have two cables going down, how do you know the distance of these two cables back to the cabinet?

Also I�m sure the estimates are using real world DSL sync speeds so could the ones further away from the pole may have a better sync speed than those closer to the pole? Might be taking into account the home wiring quality.

I agree with your thoughts on the G.fast rollout. It was supposed to be the quicker deployment over FTTP and also help FTTP as it would have required fibre lines right up to the poles or manholes.

I think from the conversations we've seen recently on here, particularly the FoD thread, the reality is there hasnt been enough fibre lines installed by OR, to the point they have had to cut back on their plans.

I have no doubt they are going full steam ahead now, with the new engineer recruitment and the opening of special fibre schools.

G.fast could still play a good role. But it needs to be the original plan of being closer to the customers.

As it stands now its being marketed to a very low percentage and I can see it being a financial loss.

I wish OR or the ISP's they feed, actually ask people what they wanted.