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 ... 35 36 [37] 38 39 ... 44

Author Topic: FTTC - Distance from the Cabinet - Realistic speed estimate  (Read 253978 times)

c6em

  • Reg Member
  • ***
  • Posts: 504
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #540 on: April 12, 2012, 08:56:05 AM »


Not as I understood it on FTTC.
I can recall seeing somwhere on Plusnet forums that the Interleaving is currently always set 'on' on FTTC and not even the ISP's can switch it off.
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #541 on: April 12, 2012, 09:30:03 AM »


Not as I understood it on FTTC.
I can recall seeing somwhere on Plusnet forums that the Interleaving is currently always set 'on' on FTTC and not even the ISP's can switch it off.

The document linked to was great for ADSL, written in easy to understand layman's terms.

However, some areas require a complete revison for VDSL2/FTTC services, such as DLM actions, BRAS Rate updates, even the use of separate modems/routers has quite an impact.

e.g. a modem re-sync does not always initiate a new PPP session, thus the IP Profile / BRAS Rate can end up stuck at a high or low level, completely unrelated to the modem's sync speed, which really confuses a large number of users as they cannot understand why their throughput appears to have no relationship to the IP Profile as reported by the BT Tester.

More often than not, the ROUTER has to be reset to force a new PPP session that will actually update the BRAS Rate / IP Profile.

Having "found out" quite a bit about VDSL2 & FTTC, mainly by trial & error, we could possibly write a similar document between us.
However, some elements of how FTTC works are apparently "secret" (especially any stats from the supplied locked modems) & the document would not be "official".


Regarding Interleaving, I am with Plusnet & their various reports (& my own logs) confirm that Interleaving can be in various states for DS & US Separately:-

OFF - Interleaving depth of 1
LOW - I have seen DS Interleaving depths of up to 1700 , still classed as LOW, while US was OFF
HIGH - I am not sure what depth signifies the start of HIGH.

Further, I have not seen MEDIUM as an interleaving state but it MIGHT exist, alternatively it may simply swith from LOW to HIGH.

Plusnet have officially confirmed to me that they have no control whatsoever over Interleaving settings for FTTC services as DLM handles it automatically.
Plusnet also confirmed they are unable to request a manual override of Interleaving settings via BT OR.

Currently, Interleaving is completely OFF for my connection (DS & US both at a depth of 1 - from my own logs).

Also this from Plusnet's reports:-

Test Outcome Pass 
Test Outcome Code GTC_FTTC_SERVICE_0000
Description GEA service test completed and no fault found .
Main Fault Location OK
Sync Status In Sync
Downstream Speed 32.2 Mbps
Upstream Speed 5.8 Mbps
Appointment Required N
Fault Report Advised N
Profile Name 20M-40M Downstream, Interleaving Off - 3.6M-7.2M Upstream, Interleaving Off
Time Stamp 2012-04-08T05:15:00


Some other documents do suggest that Interleaving adjustments CAN be requested for various profiles:-

"Standard - Auto, what we are all currently on.

Stable- Interleaving

Fast - Fastpath, lower ping for gamers."

An official FTTC specific Myths & Legends document from OR would be most helpful for end users.
However, as it might have to give some "secrets" away, I wouldn't hold your breath waiting for its release into the Public Domain.

« Last Edit: April 12, 2012, 09:32:15 AM by Bald_Eagle1 »
Logged

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #542 on: April 12, 2012, 10:01:22 AM »

@ BE,

VERY well done in beginning "our" document !

You might be interested in the news article I've just posted too !

Kind regards,
Walter
Logged

JoshShep

  • Reg Member
  • ***
  • Posts: 266
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #543 on: April 12, 2012, 10:02:35 AM »

Interesting BE, thanks for that!

JoshShep

  • Reg Member
  • ***
  • Posts: 266
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #544 on: April 12, 2012, 02:33:52 PM »

Just thought I'd update you on the issues I've been having, as it may be the same case as yours.

Had another Engineer here just 10 minutes ago, he was on the phone to the REIN team and tuned his van radio into the 612KHz frequency, drove around the street with his mobile on loud speaker to the REIN Engineer, and has confirmed there is an issue a few doors down. He has booked him to visit in the next few days. He's coming from Leeds, I'm in Barnsley so it's 30 minutes drive. Anyway he will be coming to track the REIN down and hopefully fix the problems, they can see errors on the line all the time and he said it would explain why I have a high attainable rate on 45Mbps but a lower sync on 28Mbps.

Just thought it would be possible Paul that your in the same situation as me?

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #545 on: April 12, 2012, 05:24:55 PM »

Hi Josh,

I thought your issues had been resolved by replacing the connection module at 72 m from your master socket, giving you the full fat 40Mb sync speed.

Has it all gone pear shaped again since then?

Please keep us posted regarding the REIN investigations & outcome.
It may well be worth you starting a new thread dedicated to that as a specialised topic.

Cheers,

Paul.
Logged

JoshShep

  • Reg Member
  • ***
  • Posts: 266
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #546 on: April 12, 2012, 06:01:57 PM »

Hi Josh,

I thought your issues had been resolved by replacing the connection module at 72 m from your master socket, giving you the full fat 40Mb sync speed.

Has it all gone pear shaped again since then?

Please keep us posted regarding the REIN investigations & outcome.
It may well be worth you starting a new thread dedicated to that as a specialised topic.

Cheers,

Paul.

Yeah it's all gone pair shaped again  :(

And I might as well start a few thread, I will keep you updated mate!

Cheers,

Josh

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5717
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #547 on: April 12, 2012, 07:23:30 PM »


Not as I understood it on FTTC.
I can recall seeing somwhere on Plusnet forums that the Interleaving is currently always set 'on' on FTTC and not even the ISP's can switch it off.

The document linked to was great for ADSL, written in easy to understand layman's terms.

However, some areas require a complete revison for VDSL2/FTTC services, such as DLM actions, BRAS Rate updates, even the use of separate modems/routers has quite an impact.

e.g. a modem re-sync does not always initiate a new PPP session, thus the IP Profile / BRAS Rate can end up stuck at a high or low level, completely unrelated to the modem's sync speed, which really confuses a large number of users as they cannot understand why their throughput appears to have no relationship to the IP Profile as reported by the BT Tester.

More often than not, the ROUTER has to be reset to force a new PPP session that will actually update the BRAS Rate / IP Profile.

Having "found out" quite a bit about VDSL2 & FTTC, mainly by trial & error, we could possibly write a similar document between us.
However, some elements of how FTTC works are apparently "secret" (especially any stats from the supplied locked modems) & the document would not be "official".


Regarding Interleaving, I am with Plusnet & their various reports (& my own logs) confirm that Interleaving can be in various states for DS & US Separately:-

OFF - Interleaving depth of 1
LOW - I have seen DS Interleaving depths of up to 1700 , still classed as LOW, while US was OFF
HIGH - I am not sure what depth signifies the start of HIGH.

Further, I have not seen MEDIUM as an interleaving state but it MIGHT exist, alternatively it may simply swith from LOW to HIGH.

Plusnet have officially confirmed to me that they have no control whatsoever over Interleaving settings for FTTC services as DLM handles it automatically.
Plusnet also confirmed they are unable to request a manual override of Interleaving settings via BT OR.

Currently, Interleaving is completely OFF for my connection (DS & US both at a depth of 1 - from my own logs).

Also this from Plusnet's reports:-

Test Outcome Pass 
Test Outcome Code GTC_FTTC_SERVICE_0000
Description GEA service test completed and no fault found .
Main Fault Location OK
Sync Status In Sync
Downstream Speed 32.2 Mbps
Upstream Speed 5.8 Mbps
Appointment Required N
Fault Report Advised N
Profile Name 20M-40M Downstream, Interleaving Off - 3.6M-7.2M Upstream, Interleaving Off
Time Stamp 2012-04-08T05:15:00


Some other documents do suggest that Interleaving adjustments CAN be requested for various profiles:-

"Standard - Auto, what we are all currently on.

Stable- Interleaving

Fast - Fastpath, lower ping for gamers."

An official FTTC specific Myths & Legends document from OR would be most helpful for end users.
However, as it might have to give some "secrets" away, I wouldn't hold your breath waiting for its release into the Public Domain.

Thanks for the continued education into FTTC, BE. My simple mind can't see why the same DLM logic on ADSL isn't applied to VDSL, other than for 'Secrecy purposes'. I'm not really one who subscribes to that train of thought, but take the points about the 'locked down' modems. But having a choice about whether to have IL applied or FP, should be a no-brainer, as it is with ADSL !!

From your exceptional post, it's obviously apparent that as a community we really are in the land of learning. There seems to be so many differing thoughts as to what is 'true' and what isn't.

We've had a voice-mail com from one of our senior managers today stating that 97% of Lancs will be Fibred-up by the end of 2014 (I think he said 2014 ?) thanks to the government chucking in a load of dosh, Lancs Council basically have £64 Million in the coffers now. As I've been quite busy I didn't listen intently to the message, but am sure there will be a thousand more communiques to follow on the same subject. Point being, I may just be able to give deeper 'real life' feedback as time progresses. Having said that, by C.O.P. in the year 2014, you'll probably all be VDSL scientists and have moved into trialling shape-shifting. ;) ;D
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #548 on: April 12, 2012, 08:09:49 PM »

Thanks for the continued education into FTTC, BE. My simple mind can't see why the same DLM logic on ADSL isn't applied to VDSL, other than for 'Secrecy purposes'. I'm not really one who subscribes to that train of thought, but take the points about the 'locked down' modems. But having a choice about whether to have IL applied or FP, should be a no-brainer, as it is with ADSL !!

The logic is probably very similar. I gather VDSL & VDSL2 is more or less an extension to the ADSL2+ specification.

However, purely from trial & error, it appears that IP Profile rates for FTTC connections are around 96.79%, compared to ADSL2+ at 88.2%.

It also appears that the ADSL 10 day training period is completed in either 1 day or 2 days for FTTC connections (subject to whether serious errors are detected or not).

We don't have any relevant/reliable speed v distance charts for FTTC.

VDSL2 frequencies are much higher than ADSL frequencies, with almost 4000 bit loading tones available as opposed to 216 or 512 tones.

Attenuation plays a much bigger part in the equation as frequency increases, relative to cable lengths from the cabinet.
A 5km ADSL cable length would provide some sort of connection, yet it appears that VDSL2 just cannot achieve any sort of connection over much shorter distances.

etc. etc. etc.


So yes, the logic & principles are probably very similar, but the "average" end user isn't interested in any of that.
They just expect 40Mb or now 80Mb connections (or is that now changed to 38Mb & 76Mb?)


Quote

From your exceptional post, it's obviously apparent that as a community we really are in the land of learning. There seems to be so many differing thoughts as to what is 'true' and what isn't.


So why haven't BT and/or OR released the 'truth', the whole 'truth' & nothing but the 'truth' for geeks like us?
Surely the official release of a document similar to the Myths & Legends document would quell so many rumours, misunderstandings & outright misinformation purpoted to be facts.

Quote
We've had a voice-mail com from one of our senior managers today stating that 97% of Lancs will be Fibred-up by the end of 2014 (I think he said 2014 ?) thanks to the government chucking in a load of dosh, Lancs Council basically have £64 Million in the coffers now. As I've been quite busy I didn't listen intently to the message, but am sure there will be a thousand more communiques to follow on the same subject. Point being, I may just be able to give deeper 'real life' feedback as time progresses. Having said that, by C.O.P. in the year 2014, you'll probably all be VDSL scientists and have moved into trialling shape-shifting. ;) ;D

Good luck with that & here's hoping your are closer to the head than the tail of that queue  ;)
Logged

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5717
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #549 on: April 12, 2012, 08:16:20 PM »

Ha ha, BE. Cheers for the closing comment bud. ;D

TBH, regarding a VDSL M&L Doc, I genuinely believe 'they' still don't actually know themselves ?! It's a relatively new product on our shores in relation to pumping it over our last mile of 'string', and I feel as time goes by engineer feedback will be the key to producing the said document. As somebody else commented on another thread (Iunless it was on this ??  :-[), laboratory condition testing is a world away from real-life, so I think its a case of waiting and watching BE.

Cheers.
Logged

Blackeagle

  • Reg Member
  • ***
  • Posts: 257
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #550 on: April 12, 2012, 08:50:05 PM »


It also appears that the ADSL 10 day training period is completed in either 1 day or 2 days for FTTC connections (subject to whether serious errors are detected or not).

We don't have any relevant/reliable speed v distance charts for FTTC.


My engineer told me that DLM would be active as soon as his JDSU got showtime for the first time.  He said it would have figured out the MSR 24hrs later.

With regard to distances, there is a chart here http://www.thinkbroadband.com/news/5161-how-fast-will-fttc-run-at.html  As I reckon my line is approx 650m, that fits bang on the chart with my estimate of 33Mbps download.
Logged
ASCII stupid question, get a stupid ANSI -- TalkTalk Broadband since 2006

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #551 on: April 15, 2012, 10:01:09 AM »

Update 15/04/2012:

Following the engineer's visit of Staurday 7th, apart from DLM making a couple of adjustments within the first 2 days, my connection enjoyed a period of stability.

Sync speed was around 32Mb.

SNRM rose & fell with daytime & nighttime, peaking at a slightly lower than expected 4.5dB.

Attainable rates remained steady at around 30Mb (probably a little low due to SNRM being below 6dB).

Use of the phone only lowered SNRM by 0.1db / 0.2dB.

Output Power levels stopped fluctuating.

Various error counts diminished & some remained at zero as Interleaving, INP & delay are still switched OFF.

Error Seconds still seem rather high. Increase in Error Seconds appeared to coincide with my entry onto Plusnet's 80/20 trials. As 80/20 is a complete waste of time for my connection, I have requested to be switched back to the 40 Mb service, which should happen early next week.

All in all, it really appeared that last week's engineer had completely fixed the use of the phone causing disconnections and large drops in SNRM issue.


However - there's always a however when discussing my connection - the issue appears to have resurfaced since Thursday & my connection has suffered a few re-syncs.

It is currently in snc at only 23401 k, with SNRM at 9.9dB.
Over the last couple of days, SNRM has alternated between decreasing & increasing whenever the phone is used (incoming and outgoing calls).

The re-syncs appear to have occured when SNRM tails off over a 3 or 4 minute period, a couple of times into negative values.
I have now added up to minus 5dB in the SNRM graph's Y axis.

A couple of re-syncs aren't displayed due to the scales of the graphs & the fact that the re-sync speed had only changed by a few k.

What causes the reduction in SNRM at various times of day/night, now that it is not only when the phone is used, I have no idea.
My logs don't show a sudden increase in errors to really high values (my connection has previously hung on with error bursts of 200,000 or so).

It all sounds more like a physical (still intermittent) line fault to me.

Any suggestions anyone?

Could it actually be something within the Fibre cabinet itself?


I am mindful of the as yet unknown cause of the "underlying" issue as mentioned following the engineer's discussion with OMC (Fibre centre) along with the hump in the TDR trace.

I have very recently taken to driving around with the car radio tuned to 612KHz (proper saddo) & do notice a couple of areas between home & the cabinet where the background white noise changes to a buzzing sound (during daytime).
At nighttime, the high strength of some Spanish radio station appears to block all that out at 612 KHz.

Not that it's relevant to my connection, I notice the same buzzing sound when passing through some (not all) Pelican crossings & normal traffic lights.

@BS, is that the sort of hard to describe sound?

The phone is still plugged into the test socket via a dangly filter, with all the other internal wiring completely disconnected.

The switch back to a capped 40Mb service may just have some impact, although on the 80/20 trials I did see a few days of stability, so probably not.
At least DLM won't be attempting to send anything at all to me at the highest (& most susceptible to interference) frequencies that come with the up to 80 Mb speeds.

This current disapponting position has been reported to Plusnet.

The latest 8 day's graphs are attached for reference.


Cheers,


Paul.


« Last Edit: April 15, 2012, 10:17:10 AM by Bald_Eagle1 »
Logged

JoshShep

  • Reg Member
  • ***
  • Posts: 266
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #552 on: April 15, 2012, 10:41:26 AM »

Got a few questions BE, I'm wanting to use the graphs but I have no idea how to do so. I have installed them so there not in C: Drive but what do I do now? Do I need to plug a second lan port into you router? and how do I set this up on the modem?

Thanks mate.

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #553 on: April 15, 2012, 11:09:25 AM »


Got a few questions BE, I'm wanting to use the graphs but I have no idea how to do so. I have installed them so there not in C: Drive but what do I do now? Do I need to plug a second lan port into you router? and how do I set this up on the modem?


My setup is as follows:-

Netgear WNR1000 v 3 Router, IP Address 192.168.1.254

Modem settings unchanged (apart from being unlocked), 192.168.1.1

All PCs currently connected to the router (wireless & ethernet) automatically obtain an IP Address.

Router WAN port connected to Modem LAN1 port

Spare Router port connected to modem LAN2 port.

both normal Internet access & access to the modem's stats & therefore graphing scripts are available at the same time.

The graphing scripts default to using the modem's default IP Address of 192.168.1.1
If you have changed that for any reason (some people do), the scripts will need to be edited to reflect the modem's new IP Address (search & replace is quick enough).


The script instructions that were included in the download zip file explain how to use the scripts & optionally set up Windows Task Scheduler for every minute 24/7 logging (until stopped).


Depending on your particular router, this post MIGHT be of some help:-

http://forum.kitz.co.uk/index.php/topic,10968.msg214960.html#msg214960

HTH,


Paul.


EDIT:

I would just add that the scripts do NOT work with file and folder names with spaces in them.


« Last Edit: April 15, 2012, 11:15:41 AM by Bald_Eagle1 »
Logged

JoshShep

  • Reg Member
  • ***
  • Posts: 266
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #554 on: April 15, 2012, 12:45:42 PM »

Thanks a lot mate, going to get it set up now.  :)

Edit: No stats are showing in current or ongoing folders?
« Last Edit: May 01, 2012, 05:58:48 PM by JoshShep »
Logged
Pages: 1 ... 35 36 [37] 38 39 ... 44