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 2 [3] 4 5 6

Author Topic: plusnet speeds deterioriating rapidly - plusnet says no fault  (Read 37822 times)

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #30 on: September 26, 2013, 07:43:53 AM »


** HG612 stats doesnt ever show my U1 band because its too low for the graph...  I keep forgetting to ask if theres any way I can adjust the 'y' axis graphing parameters.



That's not currently a user definable option.

What is the lowest value it has ever reached?

I'll amend the code to allow for it & post an updated graphpd.exe program.

Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33879
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #31 on: September 26, 2013, 09:20:39 AM »

From a very quick look about -28.5dBm.  I should imagine shorter lines than mine may reach a bit higher?
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

plexy

  • Reg Member
  • ***
  • Posts: 115
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #32 on: September 26, 2013, 12:14:41 PM »

Hi all,
Thanks so much for continued help and its good to know that some of my results from PN have shown they have more data available than previously given out!

HGstats has ran since yesterday and produced a few interesting results.

There were two resyncs during the period - though the change in speed was so small it doesnt show up on the graph - one to 66503 (15:11 yesterday) and one to 65911 (about 3:45 am). I suppose at least its climbing, albeit slowly. No SNRM change on the DS path, but upstream seems messy at seemingly random intervals. Two of the very small US dips are down to phone calls, but the big dip around 18:00 yesterday we were all out of thehouse and everything was switched off. Not sure about the SNRM change on US from 10am today, all the kids were out and I was asleep :D

Overall I dont see anything that screams 'downstream has a problem' yet its the DS thats getting pulled back. The US remains at 20 solid, even when 'max attainable US' goes way below 20mbps, the sync still stays there. Im also a bit unsure of why the SNRM changes on US when the US line and signal attenuation remain solid stable.

Bit of an interesting one!

Im away now for a few days but will leave the stats running and check the thread here as much as I can.

BE - would be good to see if PN can pull the crosstalk test results for you to give an answer there for sure. They seem to have the data available to them!
« Last Edit: September 26, 2013, 12:16:54 PM by plexy »
Logged
--
SSE 80/20
Sky 40/10
Zen 80/20
Bonded: 200/50 (in progress)

Hardware:Hg612/HH5/Tg589vac/Linux SBC router

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #33 on: September 26, 2013, 12:47:39 PM »


Not sure about the SNRM change on US from 10am today, all the kids were out and I was asleep :D



That just looks like a gap in the data harvesting rather than anything to be concerned about connection-wise.

I'll be releasing updated versions of the programs soon, including a few bug fixes & slight tweaks here & there.
In the meantime, I have attached a recent version of HG612_stats.exe that you may wish to use in place of the existing one from the v 1.1 download.


Quote

BE - would be good to see if PN can pull the crosstalk test results for you to give an answer there for sure. They seem to have the data available to them!



Yes, I might mither them to provide that sort of testing for my connection.

Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #34 on: September 26, 2013, 12:51:40 PM »


From a very quick look about -28.5dBm.



I've now made the Y axis for DS Power & US Power include down to -30dBm.

An amended version of graphpd.exe is attached.

Logged

plexy

  • Reg Member
  • ***
  • Posts: 115
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #35 on: September 26, 2013, 01:27:10 PM »


That just looks like a gap in the data harvesting rather than anything to be concerned about connection-wise.
Yes I concur - big gaps in that time period for other things too!

Quote
I have attached a recent version of HG612_stats.exe that you may wish to use in place of the existing one from the v 1.1 download.

Thanking you kindly. Ill get that installed now!
Logged
--
SSE 80/20
Sky 40/10
Zen 80/20
Bonded: 200/50 (in progress)

Hardware:Hg612/HH5/Tg589vac/Linux SBC router

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33879
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #36 on: September 26, 2013, 02:41:13 PM »

Do you still have DSLstats running..  if so a capture of the period 6pm to 9.30pm would be worth having a closer look at.
DSLstats is good for monitoring live and 2hr snapshots sometimes show things in a bit more clarity.  HG612_modem_stats is great for graphing and some more detailed info.  I run both as they each have their own plus points.




May be a bit late now, but you say you were out at that time... is it possible that someone rang at that particular time?.   The sharp spikes you see at about 6 & 9 look very similar to what I saw with my recent line fault when ever the phone rang.
Did you ever try as I suggested earlier to observe the SNRm when the phone rang.  You can use BTs ringback service test on 17070.  Whilst there may as well try the quiet line test to see if you can hear any noise.

My line fault mostly only affected the upstream, there was some signs on the down, but nowhere near like on the up.  I have surplus SNRm though and it was only when things were pretty dire on the upstream that Id see it reflected in the down too.
I had what the BT engineer called a 'double whammy fault'.  Some of the issues - such as random upstream SNRm fluctuations - were resolved after replacing a faulty NTE, but others such as the SNRm spikes and noisy voice calls remained until a dside swap.

Quote
BE - would be good to see if PN can pull the crosstalk test results for you to give an answer there for sure. They seem to have the data available to them!

It looks like perhaps this may be a new test? Im sure that if it was available earlier then either me or BE would have seen them. 
It certainly cant harm asking..  Although Im not sure if it would discover FEXT.. which is detected and measured from the EUsite.  I suppose it depends on how advanced the tests are..  and if a signal is reflected back down the line?   
« Last Edit: October 14, 2013, 12:14:04 PM by kitz »
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33879
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #37 on: September 26, 2013, 03:25:11 PM »

I've now made the Y axis for DS Power & US Power include down to -30dBm.
An amended version of graphpd.exe is attached.

Thank you vm BE appreciated.  Installed and run :)


Noticed that I have periods where its even lower, so I dug out one of the pbParams during that time and notice circa -36dB.  Ive still not looked to closely... but my old line, before the fault must have been really good in its heyday.   The old one had 10Mb/4.5Mb more sync in it..   whilst still running on less power  :o :o

What I do find interesting now they are graphed, is that U1 is where my power was cut back the mostest.... and guess which band I saw the most fluctuations in?
and something else I just noticed after posting the pbParams...  that was during the period when I couldnt get more than 15Mb on my upstream...  but look, my U1 power was cut back so much.. I had 0dB SNRm.   
Why the hell would the DSLAM cut my power back that much in U1 band to the point that SNRm was so low it affected my sync speed.  Something is not right there - power cut back is just not supposed to work like that.. its supposed to ramp up the power when SNRm gets low, not take it lower  ???




------------------

PS no need to recompile..  but just making you aware that there will be lines out there capable of running even lower.  Im not really that close to the cab - about 350m iirc.  I notice plexy's power in U1 is also currently running at about -32dBm


PPS Plexy,  sorry about the digressions in your thread, but its only since the advent of BaldEagle Scripts & Erics DSLstats, that we are beginning to see and learn new things about FTTC.
« Last Edit: September 26, 2013, 03:38:14 PM by kitz »
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #38 on: September 26, 2013, 10:08:08 PM »

sadly with crosstalk I think noone is ever 100% convinced, its just theories.  Like you said until vectoring is rolled and if we see our speeds suddenly jump up by huge amounts or not we dont know.

Even that engineer I spoken to near my cabinet last week wont put my issues down to crosstalk or external noise he isnt sure which.  Although I have feeling its crosstalk in my case given I had speed drops that were linked to new installs.

Vendor documentations state vdsl2 losing approx 40% of its max speed to crosstalk is 'not unusual' even in one large jump, they say if a line is next to a major disturber and that disturber gets vdsl enabled then it can lose huge amounts of sync speed in one go which is why pairs at the edges of bundles rather than the middle have an advantage with crosstalk.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #39 on: September 26, 2013, 11:56:54 PM »


<snip>

You can use BTs ringback service test on 14070.  Whilst there may as well try the quiet line test to see if you can hear any noise.

<snip>


b*cat knows that you intended to type 17070 . . .  ::)
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.

NewtronStar

  • Kitizen
  • ****
  • Posts: 4898
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #40 on: September 27, 2013, 07:05:10 PM »

This Crosstalk anomaly seems to be getting more common with VDSL2 than ADSL it must down to the higher frequencies when using VDSL on old copper lines and from I have read and are twisted pair was not designed for this, yes I am sure openreach did alot of tests before FTTC was rolled out but did they take into account that crosstalk may be introduced without significant shielding ?
« Last Edit: September 27, 2013, 07:15:41 PM by NewtronStar »
Logged

plexy

  • Reg Member
  • ***
  • Posts: 115
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #41 on: October 01, 2013, 12:28:04 PM »

Hi all,
back now from a break. The stats ran till the 29th, so ive attached the full monty graph. We wer eout of the house during this period, I left only the downstairs AC circuit on (everything else off at the breaker board) - literally only the FTTC modem and my router were plugged into AC (well, the dishwasher was plugged in but isolated by the switch on the wall).

Quote
Do you still have DSLstats running

For some of the early graphing attached here, no. From the 26th to the 29th dslstats ran happily also so I have that dtaa on a 2 hour graph cycle.

Quote
sorry about the digressions in your thread

No problem at all, its exciting stuff. Debugging and fixing problems is my hobby and also what I do for a living ;)

Quote
Did you ever try as I suggested earlier to observe the SNRm when the phone rang

Yes, I mentioned it earlier buts its probably buried in the masses of info and questions from me ;).  It does dip, but only the US SNRM. DS SNRM doesnt appear to change for a call. Its the same for inbound and outbound calls, and seems to start when ringing starts (for call termination) or when going off hook (call origination) - though the sampling rate was down to 15 seconds, i would have liked more granularity to guarantee the dip starts on ring. T

hough upon looking at the US SNRM graph, the calls are the tiny dips - not sure what caused the massive dips as they dont appear to be calls

cheers!
« Last Edit: October 01, 2013, 12:31:51 PM by plexy »
Logged
--
SSE 80/20
Sky 40/10
Zen 80/20
Bonded: 200/50 (in progress)

Hardware:Hg612/HH5/Tg589vac/Linux SBC router

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #42 on: October 01, 2013, 07:45:51 PM »

I think you require the services of an Analytical Eagle::)

Now where could he be?  :-\  Subjected to the dominating, steely glare of Mrs Eagle, perhaps?  :-X
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.

NewtronStar

  • Kitizen
  • ****
  • Posts: 4898
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #43 on: October 02, 2013, 09:28:01 PM »

I think you require the services of an Analytical Eagle::)

Now where could he be?  :-\  Subjected to the dominating, steely glare of Mrs Eagle, perhaps?  :-X

I was admiring Plexy's Graphs with interest most of it looks really good except for those US SNRM downward spikes on U2 and lessor on U0 and U1 looks fine.

but if you look you can see the US error seconds count also spike at the same time as the US SNRM
« Last Edit: October 02, 2013, 09:38:35 PM by NewtronStar »
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: plusnet speeds deterioriating rapidly - plusnet says no fault
« Reply #44 on: October 02, 2013, 10:37:13 PM »

There's also an unusually high amount of US bitswapping going on too (almost completely masking the DS bitswapping), along with a fair amount of US RSCorr errors, suggesting some sort of general & continuous US interferenece (despite US Interleaving depth being 1 - i.e. fastpath).

DS sync speed also seems a shade low, possibly due to the relatively high DS Interleaving depth of over 1000.

I'm afraid I have no idea what could be causing the larger US SNRM dips, other than possibly something being switched on/off, possibly even in a close neighbour's house?

Is there any evidence of anything untoward in the ROUTER's logs from around the times of the larger dips?



FWIW, I prefer to plot the graphs using multiples of 1,2,3,4,6,8 days, hours or minutes as it helps to keep the times along the x axis consistent.

Logged
Pages: 1 2 [3] 4 5 6
 

anything