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

Login with username, password and session length
Advanced search  

News:

Pages: 1 [2] 3

Author Topic: Sync speed on new line  (Read 1811 times)

jaydub

  • Reg Member
  • ***
  • Posts: 167
Re: Sync speed on new line
« Reply #15 on: December 14, 2017, 03:34:02 PM »

I was slightly loathe to risk taking his thread off track, as I am not suffering from his lower than expected sync issues nor his interesting looking bit loading graph.  However, as Zico has invited me to join this thread, I have attached my QLN, Hlog, DNT per tone and bit loading graphs.  (Bit loading to follow in another reply as not allowed more than 3 attachments)

I sync at 80/20, which is no great surprise having our ECI cabinet the other side of our garden wall, however I would welcome views on my QLN graph as there seems to be a greater variation between downstream and upstream figures than most of the data I have compared against.  (e.g Chrysalis: different ISP but max sync on an ECI cab).

I'm really posting just to improve my understanding of what can be gleaned from these graphs.  If the mods want to seperate this off into another thread, that's fine with me.




 
Logged

jaydub

  • Reg Member
  • ***
  • Posts: 167
Re: Sync speed on new line
« Reply #16 on: December 14, 2017, 03:35:14 PM »

.....  and the Bit Loading graph.
Logged

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #17 on: December 14, 2017, 08:29:23 PM »

It's good either way to get a rough idea of good, bad and ugly examples of the data being posted :D
Logged
mydslwebstats: Zico

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 20614
  • Over the Rainbow
    • The ELRepo Project
Re: Sync speed on new line
« Reply #18 on: December 14, 2017, 10:03:34 PM »

jaydub -- Purely by visual inspection I can say that the QLN plot shows your circuit is operating with an elevated noise floor. There is no obvious source of interference. By eye, I estimate that your circuit's average noise floor (anf) is around -108 dBm/Hz per sub-carrier. (We would typically expect a "good" circuit to have an anf value of -140 dBm/Hz per sub-carrier or lower (i.e. more negative).)

The Hlog plot tells me that there is something peculiar about the metallic pathway making up your circuit. Notice the undulations? They are abnormal. A good Hlog plot "decays" smoothly when "read" from left to right.
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.

jaydub

  • Reg Member
  • ***
  • Posts: 167
Re: Sync speed on new line
« Reply #19 on: December 15, 2017, 12:13:21 AM »

Thanks, B*cat.

My recollection is that a few months ago my Hlog graph was quite linear apart from the downturn at the end.  Unfortunately I have only just paid mu dues, so don't have any history any further back than the 4th December, which looks almost identical.

Is this something I could or should take up with IDNet.

What impact is this likely to have on performance?

I get quite inconsistent single thread speeds on the TBB Tester.  Is this likely to due to the elevated noise floor?
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 20614
  • Over the Rainbow
    • The ELRepo Project
Re: Sync speed on new line
« Reply #20 on: December 15, 2017, 05:50:13 PM »

My recollection is that a few months ago my Hlog graph was quite linear apart from the downturn at the end.

I believe you are referring to what I describe as "tail-end droop". That is quite common on those circuits which terminate in an ECI equipped cabinet.

Quote
Is this something I could or should take up with IDNet.

No. There is nothing that they could do . . . other than offer to arrange for an Openreach visit at your expense.

Quote
What impact is this likely to have on performance?

Very little, if anything is actually noticed.

Quote
I get quite inconsistent single thread speeds on the TBB Tester.  Is this likely to due to the elevated noise floor?

No. Not at all. The "TBB Tester" is concerned with the throughput -- the broadband experience -- as provided by an ISP/CP.

The Hlog and QLN plots are telling us something about the physical infrastructure of the metallic pathway -- of the G.993.2 link (and the environment in which it is situated) -- between your CPE and the cabinet based MSAN (configured as just a lowly DSLAM).
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.

jaydub

  • Reg Member
  • ***
  • Posts: 167
Re: Sync speed on new line
« Reply #21 on: December 15, 2017, 06:31:23 PM »

Thanks for the clarification (and education), B*cat.
Logged

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #22 on: December 15, 2017, 11:32:04 PM »

Yes, thanks for the info B*cat.
Logged
mydslwebstats: Zico

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 20614
  • Over the Rainbow
    • The ELRepo Project
Re: Sync speed on new line
« Reply #23 on: December 16, 2017, 02:58:17 PM »

You are both welcome.  :)

I believe WWWombat will, when time permits, make some detailed comments about what can be deduced regarding the circuit's operational behaviour.
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.

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #24 on: December 16, 2017, 04:43:09 PM »

I'm grateful for any and all assistance but I am in no hurry so there is no need for any rush.
Logged
mydslwebstats: Zico

WWWombat

  • Kitizen
  • ****
  • Posts: 1674
Re: Sync speed on new line
« Reply #25 on: December 17, 2017, 02:50:38 PM »

Apologies... Family commitments got in the way. Hopefully I'll get a mo this evening!
Logged

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #26 on: December 17, 2017, 03:12:03 PM »

No need to apologise, you're time comes before anyone else.
Logged
mydslwebstats: Zico

jaydub

  • Reg Member
  • ***
  • Posts: 167
Re: Sync speed on new line
« Reply #27 on: December 18, 2017, 05:51:40 PM »

I believe you are referring to what I describe as "tail-end droop". That is quite common on those circuits which terminate in an ECI equipped cabinet.
Tail end droop it is.

Having disconnected and reconnected over the weekend, I know have significantly more droop than before.  Too much alcohol over the weekend?!   ::)
Logged

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #28 on: December 18, 2017, 09:15:55 PM »

Started to upload to mydslwebstats under my Kitz username Zico.

I might pick up a Pi so I can leave that uploading rather than leaving my main pc on.
Logged
mydslwebstats: Zico

Zico

  • Member
  • **
  • Posts: 45
Re: Sync speed on new line
« Reply #29 on: December 20, 2017, 09:01:32 PM »

Quick answer for now, explanation later...

Hlog shows a few signs of a bridge tap, but it doesn't look to be having much impact on the line.

QLN shows that your line is seeing a lot of "noise", which in reality is other VDSL2 subscribers. This is probably the main reason for the low downstream.

There isn't likely to be a lot you can do.

I'll post more explanation later. In the meantime, can you post the graphs for the SNR-per-tone and Bits-per-tone?

Just had a thought about the reference to a bridge tap. The neighbouring farmhouse, barn conversion and new build all have dropwires that run back to the same DP pole. Of the three properties, only the farmhouse has ever been occupied (un-occupied now for a number of months) and likely to have had a phone/broadband service. If the other two properties have never been occupied, could it be possible that dropwire has been run to them but never terminated?

I've made a tweak to the earlier posted dropwire picture.
« Last Edit: December 20, 2017, 09:03:49 PM by Zico »
Logged
mydslwebstats: Zico
Pages: 1 [2] 3
 

anything