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]

Author Topic: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%  (Read 827 times)

chuffer

  • Member
  • **
  • Posts: 34

Thanks Alex.

My interest in the speedof.me results is not based so much on the speeds achieved but the pattern shown by the graph.
I understand your comment re stopping and starting but my friends results ( regardless of speed ) show a very different pattern.

Re single thread downloads, the other tests I run all show similar speeds so I'm not sure I'm suffering singles only.

Yes, I have tried other computers and browsers, and taking my laptop to the other test site and running the 7 I get much better results.

Dslreports is one of my 7 sites, usually shows high buffer bloat and timer dropped frames, it's graphs can be a bit ragged too.
Thinkbroadband is another and the analysis page often shows high upload latency.
And my KDB show a profile with retries.

A better router with more reporting might be of help, such as those by several uses here, I'm sure more could be learnt from it.

My desktop computer, I have to use a add blocker otherwise I'm waiting ages while they all load.

The laptop used for the tests does not have the blocker installed and starting broadbandspeedchecker takes 30 seconds before the test server is located whilst they load.
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 32547
  • Trinity: Most guys do.
    • http://www.kitz.co.uk

Just checking.   
When you perform a speedtest are you using Ethernet and not wi-fi.   
Are there any other devices on your network which could be consuming bandwidth?  Don't forget about such things like streaming tv/video/music.


Can you perform a speedtest here please and attach the image results as I'd like to see single thread compared to multithreaded
http://labs2.thinkbroadband.com/speedtest/


Quote
And my KDB show a profile with retries.
Is that not re-tx which simply means that you have g.inp.   I can't ever recall a KBD saying anything about retries.

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

chuffer

  • Member
  • **
  • Posts: 34
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #17 on: September 01, 2020, 02:38:10 PM »


Kitz.
Thanks for continuing with this.

Over the weekend I was missing the last few pages from the ticket and it took a while to work out that they might have been moved, I would have expected some kind of notification but as you can see I did find the new ticket.

"And my KDB show a profile with retries" you are correct, I should have said retransmissions and not retries.

I had read the Kitz pages re DLM and recall a comment that re-tx could be  off/low/high   I took the fact that it is "on" against "off" might indicate line or other issues.

I have just found a Kitz page about retransmission and low indicating g.inp is employed.

Thinkbroadband tests........

Always a wired connection and nothing else downloading/streaming.

Your request.
http://labs2.thinkbroadband.com/speedtest/
https://www.thinkbroadband.com/speedtest/1598962583301755455
----------------
Found this
"For those who want to run a test that is more sensitive to provider congestion try this special test version."

So, while I was there.
http://labs.thinkbroadband.com/speedtest/?site=omegaUEKWiwklw392
https://www.thinkbroadband.com/speedtest/1598962949515558455
Logged

Alex Atkin UK

  • Kitizen
  • ****
  • Posts: 1524
    • My Broadband History
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #18 on: September 01, 2020, 10:13:51 PM »

My interest in the speedof.me results is not based so much on the speeds achieved but the pattern shown by the graph.
I understand your comment re stopping and starting but my friends results ( regardless of speed ) show a very different pattern.

Comparing the two graphs then yes you're right, it seems a useful tool to identify where single-thread speeds are fluctuating.  Although unless you're using the same server at the same time period, it could still be the test server experiencing problems.
Logged
Exchange: INTAKE (ECI) ISP/Modems: Zen (Home Hub 5A running OpenWrt) + Plusnet (VMG-3925-B10B) + Three (Hauwei B535-232)
Router: pfSense (i5-7200U) WiFi: Ubiquiti nanoHD

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 32547
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #19 on: September 02, 2020, 12:59:10 AM »

Those results are more interesting when you compare http1 with http6.    The line certainly seems capable but its not consistent and has random jerkiness.

The 2nd one is interesting - note how httpx6 is lower than single threading.    That doesnt happen with ISP congestion.    ISP congestion shows up as a fairly straight httpx6 and a much lower single thread.   Their pipes are larger so you don't tend to see such spikyness like that either... its more balanced in that its good during the day but consistently lower at peak.
 
Wondering if there could be some local congestion though.  Its possible that a few heavy locals could cause a hot SVLAN.   Plusnet may be able to check if your SVLAN is red for you.... but the records are not always up to date.     Your friend is not necessarily on the same SVLAN as you.       
I could believe that type of pattern if there's a few houses full of kids doing bursty activities such as streaming HD YouTube vids. Is there any time of day when it's worse?
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

chuffer

  • Member
  • **
  • Posts: 34
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #20 on: September 02, 2020, 12:50:22 PM »

I have had this issue for a while during which I have had two lift & shifts and two SVLAN changes.
Tried 3/4 routers of various make/models.

I have had several OR engineers visits and usual testing always passes 100%.

I always use the same 7 test sites, running them one after the other.
I have done this at various times of the day and days of the week.
The result of which is that the speeds are very consistent.

The only variation is, shortly after a OR visit my speeds are up by 5/6 mB but returns low after 1 or two days.
I understand they always reset the DLM on completion so g.inp would be off during this time.

Worth noting that I have spoken to a few neighbours, who are with a whole range of ISP, who all say they are not very happy with their speeds.
Two are with Sky and have been told they cannot have a 80/20 package due to network issues.

A friend, on another exchange, on BT 80/20, HH6 and aprox 240 Mtrs of cable.
I have taken my testing laptop to his network and got around 76mB across the 6 tests thus proving it as OK.

He runs the same tests for comparison.

https://www.thinkbroadband.com/speedtest/1599025306377395055
https://www.thinkbroadband.com/speedtest/1599025110246974455

Logged

chuffer

  • Member
  • **
  • Posts: 34
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #21 on: September 02, 2020, 01:14:37 PM »

Should have added...................

Would I learn anything useful if I replace my Router with something with better reporting ?.

As in the post of Johnson, twinkletoes & adslmax and others.

Any recommendations please ?.

Logged

chuffer

  • Member
  • **
  • Posts: 34
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #22 on: September 16, 2020, 12:06:28 PM »

My downloads are 5/6 mB faster for a couple of days after a DLM reset.
Is g-inp off following a dlm is reset

Would somebody clarify the comments here.............
https://kitz.co.uk/adsl/retransmission.htm

"After a DLM reset the initial profile is: -

       Downstream: Interleaving (& INP) Low.  Upstream: Interleaving Off.

If the DLM identifies that the line can support Retransmission, then ReTx Low will be automatically  applied to the downstream after a couple of days."

 I read this as g-inp if off, is that correct ?..

Logged

j0hn

  • Kitizen
  • ****
  • Posts: 3046
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #23 on: September 16, 2020, 12:08:47 PM »

Yes.
Low Interleaving is the default after a DLM reset.
G.INP takes a couple days to be added.
Logged
BT FTTP 160/30 - BQM - speed test

chuffer

  • Member
  • **
  • Posts: 34
Re: Split From: Openreach FTTC Target SNR 3dB reduced Bras IP Profile to 92.64%
« Reply #24 on: September 16, 2020, 06:29:07 PM »

jOhn.

As you know, it's not possible to get up to date info from a KDB test for 11 days after a reset.

I understand that a DLM reset will zero all the error counters and that a range of things will come into consideration as the DLM examines the line which might possibly lead it to reinstate the g-inp.

But what else ?.

If it has/did enable g-inp  it does imply  there is crosstalk on the line ?.  ( almost certainly some )

What else might there be that reduces my throughput by the 5/6 mB after a few days ?.

Or is that the g-inp overhead ?.

Clearly something, as shown by the TBB graphs, is effecting my download.

My recollection is that line data ( sync, snr, line & signal attn ) seen at the router do not change  ( by much ) after the DLM resets.
Logged
Pages: 1 [2]