General Discussion
  >> Fibre Broadband


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


Pages in this thread: 1 | 2 | (show all)   Print Thread
Standard User flipdee
(newbie) Tue 14-Jan-14 22:47:16
Print Post

Difference in Speed estimates on FTTC checkers.


[link to this post]
 
Has anyone noticed the difference in speed estimations between BT Wholesale and Sky for FTTC services?
For a friends number BT Wholesale estimate the following (sorry for the formatting):

Featured Products Downstream Line Rate(Mbps) Upstream Line Rate(Mbps) Downstream Range(Mbps) Availability Date
High Low High Low
FTTC Range A (Clean) 13.6 8.4 1.2 0.8 -- Available
FTTC Range B (Impacted) 10.4 3.8 1 0.5 -- Available

WBC ADSL 2+ Up to 2.5 -- 1 to 4 Available
WBC ADSL 2+ Annex M Up to 2.5 Up to 0.5 1 to 4 Available
ADSL Max Up to 1 -- 0.25 to 2 Available
WBC Fixed Rate 1 -- -- Available
Fixed Rate 1 -- -- Available
Other Offerings
Fibre Multicast -- -- -- Available

Sky Fibre Speed Checker estimates 5.8Mbps - 5.8Mbps saying they will start the line at 4Mbps on day one and train over the following 10days.


He currently gets a maximum of 1 Mbps cownstream on a ADSL Max product, the line length to the exchange is estimated as 2935m however when measured (cables run visably along roads) I suspect is more like 3600m.

The FTTC cabinet his line is connected to is approx 2000m away using the same distance estimate via the road.

The line follows a common route to 2 other houses at the end of my friends driveway, the estimates for the other 2 houses are identical however my friends driveway is 125m long so his line "must" be at least this longer. (Which poses the question, would he be better with an FTTC modem installed at the end of the driveway, i.e, the closest point at which he can ask BT to terminate a telephone line while remaining on his land?)

I'm not sure how many other people are dealing with FTTC in what i consider "extended" mode, as I believe towards the middle of last year BT didn't even registered estimated speeds below 15Mbps I think.
I know the VDSL2 performance over distance falls in line with ADSL2+ when reaching a certain distance, however after recently having some issues with the BT Wholesale Estimator getting it very wrong at another site i'm starting to doubt it's ability to deal with these "extended" lines.
For example:
Another friends line which is 8400m length to the exchange, is approx 3000m to 4000m from the supplying FTTC cabinet (can't be sure as the cable route is a painful one.)
The BTW checker estimates:
Featured Products Downstream Line Rate(Mbps) Upstream Line Rate(Mbps)
High Low High Low
FTTC Range A (Clean) 13.6 8.4 1.2 0.8 -- Available
FTTC Range B (Impacted) 10.4 3.8 1 0.5 -- Available

WBC ADSL 2+ Up to 1 -- 1 to 3.5 Available
ADSL Max Up to 0.5 -- 0.25 to 1.5 Available
WBC Fixed Rate 0.5 -- -- Available
Fixed Rate 0.5 -- -- Available
Other Offerings
Fibre Multicast -- -- -- Available
BET Up to 2 -- -- Available
I believe he was getting 1Mbps+ on ADSL2+ before BT did a FTTC install.
The word was BT warned my friend of the performance shortfall however this obviously wasn't made clear enough at the time as they went ahead anyway.
The actual line rates for the FTTC are 1.93Mbps Download and 2Mbps Upload.
If you download any faster than 0.4Mbps (using a bandwidth limiter) the latency falls off a cliff and makes the line unusable.

Sky Fibre Speed Checker estimates 9.1Mbps - 9.1Mbps

This I find very interesting, How can Sky have a different estimate that BT Wholesale?

Also, If both checkers estimate well above 1Mbps for both lines, how can you actually make a decision on if the service is worthwhile without just ordering it and testing, if not good hopefully being able to wriggle out?

The other thing is, surely when an active FTTC modem is connected to a line the BT network loop analysis system (the name of which currently escapes me, word begins with an E with a letter missing?) will update the checker database with accurate estimates?

This is proving to make it very difficult to recommend FTTC services to friends who are at the extreme of the distance limits.

Any thoughts or advice on the above would be greatly appreciated.

Cheers,
flipdee
Standard User WWWombat
(fountain of knowledge) Wed 15-Jan-14 18:57:17
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
In reply to a post by flipdee:
I know the VDSL2 performance over distance falls in line with ADSL2+ when reaching a certain distance, however after recently having some issues with the BT Wholesale Estimator getting it very wrong at another site i'm starting to doubt it's ability to deal with these "extended" lines.

In theory, VDSL2 is capable of working as you describe. It can tail off at extended range in a way that matches ADSL2+

However, the way it is deployed in BT's network (the way required by Ofcom) makes it impossible to achieve this. Instead, VDSL2 will stop working somewhere in the region of 2-3km.

The reason: The transmit power used by the cabinet for the lowest frequencies (ie the ones already used by ADSL2+) is deliberately turned down. This is to stop cabinet-based VDSL2 from swamping the exchange-based ADSL2+ signals. Unfortunately, those signals are the ones that would work best at extended range. The system is tuned cabinet-by-cabinet (based on the distance between cabinet and exchange), so the effect varies.

And for some reason, the upstream signal actually tends to give out first at extended ranges.

I'm not sure how many other people are dealing with FTTC in what i consider "extended" mode, as I believe towards the middle of last year BT didn't even registered estimated speeds below 15Mbps I think.

BT Wholesale would report speeds below 15Mbps on the checker, but BT Retail wouldn't sell their Infinity-branded product on such a line. Other ISPs would seel it down to 5Mbps IIRC, while BT eventually came around to selling it as some non-Infinity-branded variant.

The line follows a common route to 2 other houses at the end of my friends driveway, the estimates for the other 2 houses are identical however my friends driveway is 125m long so his line "must" be at least this longer.

I've heard that every subscriber that shares the same DP gets the same estimate, but have no idea whether that means the estimate is for a modem located at the DP, or it has an average allowance for the drop cable.

(Which poses the question, would he be better with an FTTC modem installed at the end of the driveway, i.e, the closest point at which he can ask BT to terminate a telephone line while remaining on his land?)

If he could get power out there, and terminate it (and the phone line) in a suitable enclosure (which may require planning permission), then he might see some benefit. Ethernet is supposed to only work over 100 metres, and the constraint is a timing issue, not a power or noise issue. However, you might find it will still work.
Standard User flipdee
(newbie) Thu 16-Jan-14 10:48:23
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: WWWombat] [link to this post]
 
Hi WWWombat,
Thanks a lot for your reply and insight into this.
I'd love to have an arrangement with BT for telephone pole rental allowing me to extend the fibre reach (potentially only feasible by wireless) to the customers who are in this position.
My friend with the long driveway ordered sky fibre unlimited yesterday (was an o2 broadband customer so 12 month fibre with free line rental - £20 per month, would be even better if be was destined to get more than 6Mbps)
The estimator using the dp as the point of termination + an average tail length to nte makes sense from a mass calculation point of view but for accuracy in this case I'm afraid it could be a problem.

On the basis of dp as the end point for calculation for my other friend with the unworkable fttc his shared dp is actually about 1500m across two fields split at a nearby farm.
This however really screws up the estimate verses reality end result as the estimator may as welll have said no chance at the start.

I suspect the sky fibre setup will be a self install so I can hopefully do some jiggery pokery with line length in order to max out his speed.
This makes me feel I need to purchase a Fritz!box to get a better idea of line length.
Thanks for your input on this.
Cheers,
flipdee


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

Standard User flipdee
(newbie) Wed 29-Jan-14 01:36:52
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
Alright,
Done some painstaking calculations based on some real world scenarios.
BT's line length estimations (which do appear to estimate to the DP rather than the NTE within the building) against some real world FTTC speeds.

These are all connected to the same cabinet.
Estimate (see above table for headings)
FTTC Range A (Clean)13.6 / 8.4 / 1.2 / 0.8
FTTC Range B (Impacted)10.4 / 3.8 / 1 / 0.5

The actual line rates for the FTTC are 1.93Mbps Download and 2Mbps Upload.
Line length to cabinet appears to be approx 4300m (this would explain why FTTC just isn't working on this line)

It's really about 1300m to the cabinet that decent speeds are attained.
BT Estimate line length to cabinet - 1262m for 2 houses served from same DP
Downstream: 21.86 Mbps
Upstream: 20 Mbps

Downstream: 21.53 Mbps
Upstream: 10 Mbps

(Second house doesn't subscribe to Infinity 2 so slower upstream.)

Now the question is, for my other friend who's getting self install sky fibre activated on friday?
The annoying thing is, the BT estimator seems to be broken.
A line length to the exchange (from a house with the FTTC cabinet on their door step gives 3350m but an estimate to the exchange from his house gives 2935m!

But he's at least
3593.60m distance via road google maps to exchange
1979.95m distance via road google maps to fttc

Bah bah bah, I suppose i'll just have to wait and see!
If only there was a line length estimator available for FTTC, this "impacted and clean" rubbish is nearly useless, at least with actual line length to the exchange it would give a fairly good idea.

Cheers,
flipdee
Standard User flipdee
(newbie) Wed 29-Jan-14 02:08:03
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
Quick question, does anyone know if a checker out there at the moment will flag up fttc on a line already?
Would an minimum upstream profile always be 2mb on fttc?
flipdee
Standard User RobertoS
(sensei) Wed 29-Jan-14 02:21:29
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
Your upstream line speeds look like the IP Profiles. Upstream ones are always set to the number in the product spec, and are nothing at all to do with the connection speed. Though the connection speed cannot exceed it.

I see no logic in those downstream distance and speed figures. I think you may be getting some wrong or incompatible info from somewhere.

If your friend getting Sky fibre currently has a Sky LLU phone, (Sky Talk), then the BT Wholesale checker has never worked with those on ADSLx, nor with TalkTalk LLU phones. By definition it reports on lines it can access. It can't access LLU ones. In fact, OfCom regulations would prevent it doing so even if it could.

Most people on these forums with FTTC who have commented on the recently introduced clean and impacted figures have said how close they are to their actual figures.

My broadband basic info/help site - www.robertos.me.uk | Domains,site and mail hosting - Tsohost.
Connection - Plusnet UnLim Fibre (FTTC). Sync ~ 59.4/14.4Mbps @ 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.
Standard User flipdee
(newbie) Wed 29-Jan-14 08:31:42
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: RobertoS] [link to this post]
 
Hi RobertoS,
Thanks for your reply.
That's what I thought, The reason why these downstream speeds are weird is down to the fact the lines are too long to support anything higher.
I don't know what the "safe" minimum in FTTC is, but these lines I estimate to be 1600/1700m long. Ironically if the speeds listed are stable then this will be at least 4 times quicker than the ADSL Max/2+ speeds are capable of on these lines.
My friend getting Sky fibre was on o2 phone and broadband but not their LLU product, i've noticed the address checker matches up with the actual phone line checker results for several numbers i've checked, the line length estimates break once you pass the FTTC cabinet by about 3 or 4 hundred meters, I don't have enough actual telephone numbers for the area to make a definite conclusion.
That's a great tip on the upstream profile, means both of these lines are definitely not ADSL2+.
So we're probably looking at a real world maximum FTTC line length of 1500 to 1600m, amazingly BT seem to supply people with a line longer than this, in my friend's ~4000m to FTTC cabinet length BT did warn him about it not being great, however as I said in my OP syncing at 1.98meg is not the same as actually carrying that speed.
I reckon 0.6meg is about all it's capable of.

Obviously this scenario is pretty unusual (although I've got about 10 friends so far in this boat).

Cheers for your help,
flipdee
Standard User WWWombat
(fountain of knowledge) Wed 29-Jan-14 15:03:50
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
Remember, length isn't everything. Diameter matters, as does the material the wire is made of.

Thicker cables can get longer range than thinner ones, possibly significantly so. Aluminium cables get less range.

Standard wire is 0.5mm. IIRC BT have been known to use 0.7mm and 0.9mm for longer lines. And thinner wire, in places, but not normally on longer lines.
Standard User RobertoS
(sensei) Wed 29-Jan-14 16:19:31
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
In reply to a post by flipdee:
That's a great tip on the upstream profile, means both of these lines are definitely not ADSL2+.
Note that was in relation to FTTC IP profiles, as FTTC is what the thread is about.

I can't remember what happens to ADSL2+ upstream IP Profiles, but in any case the maximum upstream sync on Annex A is 1.3Mbps, maybe 1.4Mbps, and on Annex M 2.5Mbps. To get 2.5Mbps the line needs to very short - you are thinking in terms of short enough to hit 20+Mbps downstream. On downstream sync's of below 16Mbps Annex M is counter-productive and should not be used.

My broadband basic info/help site - www.robertos.me.uk | Domains,site and mail hosting - Tsohost.
Connection - Plusnet UnLim Fibre (FTTC). Sync ~ 59.4/14.4Mbps @ 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.
Administrator MrSaffron
(staff) Wed 29-Jan-14 21:29:02
Print Post

Re: Difference in Speed estimates on FTTC checkers.


[re: flipdee] [link to this post]
 
At the end of the day the only real method is a VDSL2 modem with access to stats on a line.

Too many variables to expend too much time on guess work

Andrew Ferguson, andrew@thinkbroadband.com
www.thinkbroadband.com - formerly known as ADSLguide.org.uk
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Pages in this thread: 1 | 2 | (show all)   Print Thread

Jump to