Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: 1 ... 4 5 [6] 7 8 ... 44

Author Topic: FTTC - Distance from the Cabinet - Realistic speed estimate  (Read 254033 times)

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #75 on: August 10, 2011, 09:19:46 PM »

Walter

No mate, once we've tested that we are receiving the 40/2 or 40/10 at the PCP, we connect through the old/existing E-side carrying the dial tone, pick that and the combined DSL signal up in the 'return' cable and put it through to the on-going D-side. (As I'm sure you already know Walter, just info for those who don't).

After leaving the PCP and visiting the EU's, all testing is carried out at the premises, ergo no disconnecting of the E-side.
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #76 on: August 10, 2011, 09:36:48 PM »

BE .......... reading your post, you are now well-versed in the shennanigans that make up DSL mate  ;D.

All that is correct, barring possibly the comment about SNR. That is all dependant on 'factors'. Your SNR of 5.9dB on the DS is pretty much perfect, if it was higher, I would suspect some kind of fault-type condition. However, if you live very close to the FTTC a SNR of 15dB is also very good. I think Jeff is the man to explain it if you need a more in-depth angle on it. :).

I genuinely cannot offer up any more advice that may help you. I'm as baffled by this as you are ?? I can't accept that you have gone from 35 Meg stable synch to 22 Meg via DLM !!!! It's just too large a jump imo.

Having read your post though, you put it quite succinctly as to how the DSL drops over attenuation (in laymans terms that is ....... the real reason is that it drops due to the ammount of 'noise' spread over the frequency bins, and how much data it can load in each bin), and it really is as simple as that, the longer the line the higher the attenuation !! I was maybe over-thinking the situation in my earlier post. Your line basically 'starts' at the FTTC now, so your attenuation should be low for a D-side of just 0.5 Miles ??!! As I say, something doesn't quite add up. :-\
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #77 on: August 10, 2011, 09:39:50 PM »

Just seen your 'edit'.

That is the only thing that makes sense, that there is a 'cap' applied. But then again, I would expect to see a lot higher SNR if that was the case, roughly 20dB and above.
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #78 on: August 10, 2011, 09:43:23 PM »


Having read your post though, you put it quite succinctly as to how the DSL drops over attenuation (in laymans terms that is .......


@ razpag,

Sorry about the laymans terms. I didn't want to confuse you with a load of technical jargon.  :lol: :lol:

Paul.
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #79 on: August 10, 2011, 09:46:55 PM »

You jest sir, but with your learning accumen you'll probably be at BT Martlesham next week with our boffins !!! ;)
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #80 on: August 10, 2011, 10:12:18 PM »

Just seen your 'edit'.

That is the only thing that makes sense, that there is a 'cap' applied. But then again, I would expect to see a lot higher SNR if that was the case, roughly 20dB and above.

I'll ask Plusnet tomorrow about a possible cap at their end. If I'm thinking correctly, that wouldn't cause any effect on SNR, would it?
With Plusnet, a manual adjustment to user profiles has to be made after installation of FTTC.
e.g. When mine was installed, I had to wait for around 1 hour before I got the full whack that my line could deliver.
That was 32-33 Mb, just in case I haven't mentioned it before ::)
I was just about to phone Plusnet to ask for a profile nudge when they phoned me to say they had just nudged it.

When the BT modem was replaced last Thursday, BT said they had reset my profile at their end to the Max & if that was only 24 Mb at my end, it was because that was all my line could deliver.

It may just be that Plusnet now have to manually reset my profile back to the Max at their end.

That sound too simple though, so I'm not putting a week's wages on it being the answer.

Paul.
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #81 on: August 10, 2011, 10:33:26 PM »


I can't accept that you have gone from 35 Meg stable synch to 22 Meg via DLM !!!! It's just too large a jump imo.


I still don't know whether or not interleaving affects line speeds & I can't find a clear answer anywhere. I believe it affects download speeds though.

At one stage Plusnet informed me that interleaving was automatically set at HIGH for downstream & ON for upstream due to the disconnections.

Is interleaving a separate stabilisation measure from DLM, or is interleaving the effect of DLM?

Paul.
Logged

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #82 on: August 10, 2011, 11:59:08 PM »

A Paul & RP,

Well done for summarising things so well.

A small point re attenuation has been hinted at Viz the attenuation value varies with the tones (Frequencies) involved.
(E.g. with the 2Wire it reports an attenuation at 300 kHz as well as what I understand to be an average attenuation, which is usually a little lower, over the entire frequency range used.)

I assume that on an exchange based service the JDSU attenuation calculation must observe the whole line length and the front end components on the line card etc within the exchange.
Similarly with a FTTC based calculation there must surely be the line card com ponents within the FTTC DSLAM as well as perhaps the remaining voice component back to the exchange if the filter cct. is misbehaving. However in both cases the active frequency range starts at tone 6 for the upstream.

Another point is that just as there is often a difference in the attenuation reported on the same line when it changes from ADSL 1 to ADSL 2, the same logic must apply when it becomes VDSL2 as the potential frequency range has increased. Thus you might expect a much larger variation in attenuation if going from a long ADSL1 line to a much shorter VDSL2 line.

How all this might reflect into operational maintenance and fault finding procedures is something I doubt if anybody here can tell us yet. In an ideal world it would be very valuable to change Paul's FTTC DSLAM port but I very much doubt if that has been developed in the new FTTC engineers' bible ! In any case that would presumably involve instructing the DSLAM to change the active port for the particular cct ID.

Can anybody else add more to this please?
__________________________________

See also Kitz's explanation

 ~ Tones which aren't in use.

Certain sub carrier channels are not used. Some of these are laid down in the g.DMT standard, whilst some others may depend upon the DSLAM/MSAN manufacturer and vary slightly.
Common tones not in use are:-

DC (First Tone). Tone 0.
Guard Band (Tones 1 to 5 < 25.875 kHz). Tone 1 POTs. Tones 2-5 prevents cross talk between POTs + adsl.
Guard Band (Tone 32 - 138kHz). Prevents cross talk between upstream and downstream data.
Nyquist frequency (Final frequency tone)
Upstream Pilot Tone. (Tone 16 - 69kHz)
Downstream Pilot tone. (Tone 64 - 276kHz).
Annex_M Stop Band. (Tone 59 - 254kHz). ADSL2+ Annex_M only.
DSLAM Specific Tones (eg Tones 476 - 499 2053kHz-2156kHz on Be*/02 MSANs only).
Your router will also mark any sub-channels where the SNR is too low to carry data as unusable.

Read more: http://www.kitz.co.uk/adsl/adsl_technology.htm
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #83 on: August 11, 2011, 12:19:21 AM »


How all this might reflect into operational maintenance and fault finding procedures is something I doubt if anybody here can tell us yet. In an ideal world it would be very valuable to change Paul's FTTC DSLAM port but I very much doubt if that has been developed in the new FTTC engineers' bible ! In any case that would presumably involve instructing the DSLAM to change the active port for the particular cct ID.


Hi Walter,

Would your suggestion be the L & S (Lift & Shift) operation?

I was wondering if that could be an option now.
If indeed my profile is now stuck, either at BT's end or at Plusnet's end, with both of them saying they are giving me the Max Rate possible, I am sure I have read somewhere that a L & S operation can usually free things up.

I'll see what Plusnet have to say tomorrow when they can hopefully prove or disprove my claims of really decent download speeds.

If the whole matter is due to a dodgy line/connection somewhere, it probably wouldn't solve that, but at least it might be like starting from fresh again.

I do apologise for being a bit of a P.I.T.A. with all this. I am trying to learn / understand as I go along, as well as get my speeds back again.
I might even get my life back again too. This silly issue seems to have taken over a little.

Mrs. Eagle would be pleased too, because I am supposed to be refurbishing the house, not messing about with modems, attenuation, SNR & so on.

Paul.
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #84 on: August 11, 2011, 08:25:29 PM »

Hi All,

Just an update from my discussions with Plusnet today:-

There is no further progress >:(

What a crying shame I hear you say. He'll be around for a bit longer mithering us to death then.

I am now achieving 22 Mb downloads which is more than my originally estimated speed of 14.6 Mb, so tough!
That is apparently quite good for a line length of 5283 m. Does that length ring a bell Walter?
It is the figure that Plusnet quoted to me today, just to prove I have a long line & my expectations are too high

Plusnet have not capped my profile. It is currently being reported via Plusnet as 20 - 30 Mb, even though they say it is set at 37 Mb at their end.
When I mentioned that their own ticket system had previously reported it as 20 - 40 Mb, the response was that it does go up & down a bit. I fully accept that comment, but by 11 Mb or so?

Plusnet don't know or apparently care why my speeds have dropped. They were more than I should have been achieving anyway.
They are certainly a lot more than my previous 1 Mb via ADSL. I don't believe that to be a fair comparison from a technical support staff member.

I can request another engineers visit though, to do the tests that the "special" engineer should have done on Tuesday (when I waited at home all day for nobody to turn up), & again yesterday when the "special" engineer just replaced the BT modem & let me see the stats that we can't normally see because the BT modems are locked/disabled.

I may have to pay £144 + VAT for this visit though, if the engineer can't find a fault.
Well, if the engineers don't actually run the "right" tests, they obviously won't find any faults.

BT have not capped my profile either. My current 22-23 Mb speed is proof that I cannot achieve anything faster?

Do I still have an intermittent fault? I won't know until I lose connection or experince a crackly phone line again.
I suspect I still do have though as no "special" tests or work has been carried out on my line.
By "special", I mean at least the TDR test that the "special" engineer could have/should have carried out yesterday.
I had no idea at the time that the JDSU that he had in in hands could have at least carried out that test.

Has the new & improved version 2B BT modem "fixed" everything? I don't know, maybe (apart from my speeds).

Have I ever achieved decent speeds? Yes, as confirmed by Plusnet viewing my speed test in their ticket system.

Have BT ever proved that I have ever achieved decent speeds? Yes, as I confirmed in Plusnet's forum http://community.plus.net/forum/index.php/topic,96832.0.html
There it is razpag. The proof that I have searched for high & low!!!! I suspect you were beginning to think that I had never really synced at anything over 30 Mb.
It actually got better than that for a good few days. I didn't keep checking speeds then though, beacuse everying internet-wise was instant.

I might have to edit the very last sentence in that post, regarding the accuracy of Plusnet's availability checker.
What I meant at that time was at least it showed I could have FTTC, when the providers, BT, were saying it wasn't even available for me.

Can Plusnet examine their own logs of my decent connections as proof? No, they were auto deleted when BT reset my profile last week.

Do I really have a line length of 5283 m? Maybe, right back to the exchange (for phone purposes only)

What is my reported line length for FTTC broadband purposes? BT & Plusnet don't know. I estimate around 800 m or less from driving the route of the overhead cables that I can see to the cabinet (PCP 51) that BT have confirmed I am connected to.

What line length does my attenuation of 30.2 dB & SNR of 5.9 dB downstream suggest? Plusnet don't know. I can't ask BT as I'm not allowed to speak to them & I don't know how it is calculated for FTTC

How do BT & ISPs estimate likely speeds before signing up for FTTC? I don't know, Plusnet don't know except that "various" factors are taken into account. Maybe BT know, but are just keeping it a secret.

Do I need to consider the effect of interleaving now? Not at the moment. I have seen today's Plusnet report that confirms it is OFF for both uploads & downloads.

My neighbour, in the other half of the semi that I live in, does NOT have FTTC installed. Ah, good, what line length is reported for her ADSL connection? 5283 m. Well at least that proves that BT & ISPs are using irrelevant figures for estimating FTTC speeds, or does it? It is the length that Plusnet quoted to me today, so it must be, musn't it?

Is this whole thing an exact science? It would appear not at the moment, at least as far as those providing the service between them are concerned.

Are there many others in a similar position to me? I don't know, but I would imagine so, apart from those who live next to the FTTC cabinets.

Will this thing ever get resolved? Hopefully, given time.

Is it really the big deal that I have made it out to be? I think so, a loss of 10 - 12 Mb download speeds seems too significant to ignore.

Are download speeds guaranteed & stable. Absolutely not.

Why not? Too many variables. Even someone switching on a hair dryer or their flashing Christmas tree lights next door can have an effect.
Does my neighbour switch hers on a stupid o' clock in the morning & cause my disconnections? I have never heard the hairdryer & it's the middle of summer FGS.

Is this a waste of time? I hope not for my personal case & I hope that once I have all the answers it may be of some help for others in similar situations.

Is there a reliable database of FTTC real user achievable speeds & connection statistics? I have not managed to find one. I might start one off with a link to Kitz.co.uk.
Obtaining connection stats is probably the biggest hurdle, unless you collar the friendly engineer when he installs FTTC, or turns up to "repair" your faults.

Why are the BT modems disabled for obtaining connection stats. I don't know. Maybe it is just to make it too hard to prove you have a sub-standard connection?

Thanks all for tolerating this ranting & rambling from me (for now).

Paul.
« Last Edit: August 11, 2011, 09:37:43 PM by Bald_Eagle1 »
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #85 on: August 11, 2011, 09:36:32 PM »

Mr Eagle -- My answer to one of your earlier questions is that interleaving only affects the speed of data throughput and not the line sync speed.

My comment on your situation, having re-read everything you have written, is that you have an intermittent fault on the copper pair, which manifests itself as a crackly line. Your voice line, thus, is not operating within specification and your telephony provider is required to correct the fault. To do so, they must employ a representative of the owner of the infrastructure to trace the fault and then fix it.

Your telephony provider = the company you pay for your telephone calls.
The owner of the infrastructure = OpenReach.

Once the underlying copper pair fault is correctly fixed, then we should review your broadband (FTTC) speed.

Finally, have you made direct contact with Bob Pullen of PlusNet?
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #86 on: August 11, 2011, 10:28:21 PM »


Finally, have you made direct contact with Bob Pullen of PlusNet?


Hi burakkucat,

I have tried to contact Bob via Plusnet's service.customer-feedback usenet group (subject F.A.O Bob Pullen - Ticket #44208926) a few days ago as many others appear to achieve success via this route.

Unfortunately I have not received any response from him yet.

BTW, I pay Plusnet for both my phone & broadband services since my switch to FTTC.
Prior to that, I paid BT for my telephone & Plusnet for my broadband.

I have been a Plusnet customer in its various guises (Free-Online, Force9, Plusnet) for many years & have previously been very satisfied with their customer service & fault resolution processes.

Indeed, I am not actually complaining as such now (although it may be coming across that I am). I am just becoming a little more frustrated each day with responses that don't really appear to address the underlying issues.

I am more than willing to accept that FTTC is a reasonably new technology that is not fully understood at this time, even by those who deliver the service i.e. BT & ISPs between them, as long as they don't give up and/or hide behind inaccurate & very conservative speed estimates.

Paul.
« Last Edit: August 11, 2011, 11:14:14 PM by Bald_Eagle1 »
Logged

jeffbb

  • Kitizen
  • ****
  • Posts: 2329
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #87 on: August 11, 2011, 11:20:55 PM »

Hi
quote Signal to Noise Ratio also affects line speed. A high SNR = good, a low SNR = bad.

Just to clarify

SNR  Signal to Noise Ratio  the higher the better ,that is like if 2 people are talking (that is the signal) then the lower the background noise the better
.
But There is what is known as  SNR Margin normally 6db is recognised as the  minimum amount of Spare SNR between the level of line noise and the transmitted signal(normal target SNR) . In the above analogy would be at what point would background noise begin to affect the 2 people talking .So immediately after synching you would expect to have a margin of about  6db.
Here HIGH SNR margin is not good .
Normally the Higher the SNR margin the lower the Synch Rate .
High Target SNR
Any connection is monitored and its quality is measured .Where a line is found to be unstable then the  target SNR nominally 6db is increased ,this is in 3db steps up to a maximum of 15db.

on 8max this can cost about  800kBps per 3db above the nominal 6db. Depending on Bit allocation .The more tones used then the greater the effect of increased SNR margin.(about 4Kbps/db/tone) .

After  a connection is made then the SNR margin invariably changes with time . Gradual changes are normal during any 24hour cycle . sudden increases indicate abnormal noise conditions .

Funnily enough this is where Higher snr margin indicates an improving line condition whereas a drop in SNR margin indicates a worsening line condition
Regards Jeff


 


 
Logged
zen user

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #88 on: August 11, 2011, 11:38:49 PM »

@ jeffbb

Thanks for that info.

I see you are currently experiencing your own SNR / SNRM issues that have affected your sync rate in quite a big way:-

http://forum.kitz.co.uk/index.php/topic,9752.0.html

Paul
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #89 on: August 12, 2011, 08:17:40 PM »

Quick Update

Modem No. 3 (version 2B) has surprisingly not "fixed" everything for me.

Suffered just 1 disconnection today, the only one since Wednesday, but the DLM still robbed me of 6 Mbps or so.

Downloads now back down to around 17 Mbps.

Good news: One of Plusnet's staff appears to be taking a slight interest. It is the same guy that opened my original ticket back on 5th July.

http://community.plus.net/forum/index.php/topic,97734.0.html if anyone has the slightest interest. My bit starts quite a way down the post.

The ping looks O.K. though, not that it really matters to me as I don't play online games.

Paul.
Logged
Pages: 1 ... 4 5 [6] 7 8 ... 44
 

anything