Kitz Forum

Broadband Related => FTTC and FTTP Issues => Topic started by: AArdvark on April 15, 2015, 12:58:44 AM

Title: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 12:58:44 AM
BE1,

Please explain this,  it does not fit your figures.
(Before I ask Plusnet I would like to know the Correct answer  ;D )

FYI:
Before my current issues with Plusnet, I had a 'Stuck' profile of 68M (According to Plusnet's tests)
The OR Engineer was supposed to do a simple DLM Reset to clear the stuck profile.
I appear to be at the same position I started at before the last OR visit.
Please confirm or deny my reasoning, I am doubting my own thoughts now !!!
Thank You.

Sync via HG612 Stats from Zyxel vmg8324-b10a:
(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2000.48.43_zpsx4ejtado.png)

Result from BTW Speedtest:
(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2000.31.08_zpsvzimmawf.png)

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2000.32.42_zpsucsx9pvv.png)

Title: Re: Aardvarks line deterioration
Post by: kitz on April 15, 2015, 01:14:21 AM
Can I just confirm that Im seeing things right.   You have a sync of 69670 & an IPprofile of 67.44.

Yet the BTw speedtester is giving you a full 73.53 Mbps.    I take it that youre not seeing those speeds off the BTw network?



Your line is still running without any DLM profile since the last engineers reset and g.inp is still off.
Your IP-profile is 96.79% of your sync speed..  which fits with what BE1 was saying.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 01:39:11 AM
Can I just confirm that Im seeing things right.   You have a sync of 69670 & an IPprofile of 67.44.
YES
Yet the BTw speedtester is giving you a full 73.53 Mbps.
YES
I take it that youre not seeing those speeds off the BTw network?
Please explain ???
Your line is still running without any DLM profile since the last engineers reset and g.inp is still off.
YES
Your IP-profile is 96.79% of your sync speed..  which fits with what BE1 was saying.

How is it possible to get a Speedtest that is greater than your IP Profile?

This is the original fault that Plusnet called a 'stuck profile'.
This is the fault that was supposed to be fixed by a DLM Reset.
Title: Re: Aardvarks line deterioration
Post by: kitz on April 15, 2015, 02:16:37 AM
Quote
How is it possible to get a Speedtest that is greater than your IP Profile?

Its not.   

However, the BTw speedtester sits at the RAS and if you do something called a TAP3 test (see bottom of page here (http://www.kitz.co.uk/adsl/btwperformancetest.htm), then it completely bypasses your ISP... so in your case it would be testing your speed over the BTw network directly between your home and Manchester.    You will have known though if you were doing a TAP3.

Quote
I take it that youre not seeing those speeds off the BTw network? Please explain ???

As above the test sits in or on the BT Wholesale network, although you go through your isp gateway, in effect your going over the BTw network, to PN gateway and then back up the BT network again to Manchester RAS and no external Internet.

From your comments I take it that the BTw speedtester repeatedly consistently gives you a speed that would equate to 80Mbps.   That I cannot explain other than to say it is impossible or the performance tester  is broken. 
Its not possible for a line syncing at 70Mbps to carry data more than the allocated bit rate.

There is something really very odd going on with your line.  Im beginning to have a horrible niggle about overheads and separate bearer channels though how that could work Ive no idea.   
 
When you tested with the HG612
1) Do you have any full stats from xdslcmd info --stats
2) What was the maximum length of time it was connected for.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 03:14:47 AM
Quote
How is it possible to get a Speedtest that is greater than your IP Profile?

Its not.   
That is what I said to Plusnet and I sent Screen Caps although the results are recorded by BTW if anyone doubts.

However, the BTw speedtester sits at the RAS and if you do something called a TAP3 test (see bottom of page here (http://www.kitz.co.uk/adsl/btwperformancetest.htm), then it completely bypasses your ISP... so in your case it would be testing your speed over the BTw network directly between your home and Manchester.    You will have known though if you were doing a TAP3.

Quote
I take it that youre not seeing those speeds off the BTw network? Please explain ???

As above the test sits in or on the BT Wholesale network, although you go through your isp gateway, in effect your going over the BTw network, to PN gateway and then back up the BT network again to Manchester RAS and no external Internet.

From your comments I take it that the BTw speedtester repeatedly consistently gives you a speed that would equate to 80Mbps.   
Have result from original OR Call and I have just run it 3 times and got results in the 70's

That I cannot explain other than to say it is impossible or the performance tester  is broken. 
Its not possible for a line syncing at 70Mbps to carry data more than the allocated bit rate.
I know. As my stats are ignored by Plusnet, I am waiting to see if they are willing to say anything about BTW's own results.
There is something really very odd going on with your line.
I believe the phrase is 'No S**t Sherlock !!!' <jk>. It is encouraging that someone else is convinced, just now need to make that Plusnet and I will be Happy.

Im beginning to have a horrible niggle about overheads and separate bearer channels though how that could work Ive no idea.   
Nicely leads to my question regarding G.INP disappearing which got a 'Not in my remit*' answer from Plusnet, along with the other 5 Questions. [*That is a quote]

When you tested with the HG612
1) Do you have any full stats from xdslcmd info --stats
Looking now ..... will not be much.
2) What was the maximum length of time it was connected for.
Not long as the line was 'screwed up' by OR ,it did not seem to matter testing the HG612 and I got a 5M higher sync with the Zyxel so it was swapped back.

Additional Results from BTW Speedtest:

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2002.36.08_zps0i89bpu7.png)

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2002.47.44_zpsyht64bnf.png)

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-15%2002.48.33_zps46okwiqj.png)

Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 03:31:24 AM
Missing answer from previous Reply:

I am loath to do the TAP3 test as the last time I had great problems getting the PPP session to authenicate when I put the correct UserID/Password back.
This kicked off DLM because of all the bouncing of the line.
(Had to power-off a lot of times to try getting the Plusnet PPP session to work with the HG612 then switch back to the Zyxel which did not work.)
If it is useful I will do it.  :fingers: :fingers: :no: ;D
Title: Re: Aardvarks line deterioration
Post by: Bald_Eagle1 on April 15, 2015, 05:41:32 AM
I can only think that the BT test is wrong.

As mentioned by kitz, your IP Profile looks correct for the sync speed & it is physically impossible for throughput to be higher than sync speed.

What is the Current Line speed reported by PlusNet in the member centre?


Title: Re: Aardvarks line deterioration
Post by: kitz on April 15, 2015, 10:30:39 AM
Missing answer from previous Reply:

I am loath to do the TAP3 test as the last time I had great problems getting the PPP session to authenicate when I put the correct UserID/Password back.
This kicked off DLM because of all the bouncing of the line.
(Had to power-off a lot of times to try getting the Plusnet PPP session to work with the HG612 then switch back to the Zyxel which did not work.)
If it is useful I will do it.  :fingers: :fingers: :no: ;D

No dont bother with a TAP3,  I dont think it would provide us with any useful data.
I wouldn't mind seeing a TBB speedtest though to look at httpx1 v httpx6.

Your upstream throughput seems low - is it always around 15Mbps?   I'd expect that to be somewhere in the region of 18.5Mbps for a 20Mbps sync.
Also can you provide me with a tracert to the BBC.   Ive noticed a couple of those performance tests show very high ping rates - but that isnt conclusive as it is known to often be out when it comes to ping time.

I know what other test is at the back of my mind and I know that you wont like it :-\  Im also afraid to suggest it in case it does make things a lot worse and you will likely lose some speed.   :paperbag:
At least your line has been stable for the past 10 days, but its also strange that it seems completely free of any DLM.   A resync could change all that.
It also concerns me that since around the time g.inp was enabled, you have had episodes whereby both the HG612 and Zyxel have problems obtaining authorisation.   I know that Ive already checked with you that the HG612 was on SP08.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 12:29:17 PM
I can only think that the BT test is wrong.

As mentioned by kitz, your IP Profile looks correct for the sync speed & it is physically impossible for throughput to be higher than sync speed.

What is the Current Line speed reported by PlusNet in the member centre?

67.4Mb

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-03-24%2003.30.38_cut_zpsysutzykl.png)
Title: Re: Aardvarks line deterioration
Post by: Al1264 on April 15, 2015, 12:40:02 PM
Could you have typed the wrong phone number into the (BTw) speed tester (very unlikely, I know)?
Could the BT database be mixing your line with another (possible?)?
The reason for these questions is, if you run a BTw speedtest on one BTw line but enter the phone number of another you can exactly replicate what you're seeing (measured download speed from a 'better' line exceeding the reported IP profile for a 'worse' line).
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 15, 2015, 12:51:31 PM
Quote
Could you have typed the wrong phone number into the (BTw) speed tester (very unlikely, I know)?
;D ;D ;D ;D
I am sure it was the right number, also it is remembered in Firefox and I can recall the last entered value for the field.

Quote
Could the BT database be mixing your line with another (possible?)?
I have no way of knowing but I do doubt this, BTW/OR are not that bad ........ surely.  ;D

Title: Re: Aardvarks line deterioration
Post by: kitz on April 15, 2015, 01:02:32 PM
It allows you to do a speedtest and gives you the results in text format before you enter the phone no.
Also because the IP profile is 96.79% of his sync speed, I dont think in this particular instance it is getting info mixed with another line.
Title: Re: Aardvarks line deterioration
Post by: Al1264 on April 15, 2015, 01:45:09 PM
Yes, sorry, missed the sync confirmed from the router.  :-[
Title: Re: Aardvarks line deterioration
Post by: WWWombat on April 16, 2015, 12:30:45 AM
This is the original fault that Plusnet called a 'stuck profile'.
This is the fault that was supposed to be fixed by a DLM Reset.

The data in the original screenshots don't agree that this is a "stuck profile". The profile matches the sync speed perfectly, so I don't think this is the issue.

The real problem is that the speed displayed in the browser, after running different flash-based applets, suggest something impossible. A speed that has defeated two speed control mechanisms (and the PN one is a very active mechanism) and beaten the physical restriction of the sync speed.

If we do a Sherlock - and eliminate the impossible - then it suggests we are being given incorrect data in one or more place.

At this point, I would take things back to basics.

First would be to ensure I was connected as expected (eg not using the neighbour's WiFi accidentally, etc) with plain ethernet cable, and wireless totally disabled.

Second would be to confirm the full sync data from the modem - ie plain text from a telnet session, or by running the snapshot graphs, and copying out of the Plink file.

Third would be to run a download as raw as possible, and time it manually. I would run FTP or WGET from the command line under Linux (or Cygwin under Windows), but I'm sure there are plenty of other options.

I'd then repeat on other machines, if possible.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 16, 2015, 09:50:57 AM
Quote
The data in the original screenshots don't agree that this is a "stuck profile". The profile matches the sync speed perfectly, so I don't think this is the issue.
Plusnet's definition not mine. I am not allowed to question it or anything, so it seems.
Ditto for anything OR derived which is the point of the question. :)
I have so far been told that the Plusnet RADIUS graphs are wrong but not to worry about that.
Now BTW/OR Speedtests are unreliable, so why the unbending reliance on tools that do not work. ???
Why the inability to question what is 'impossible' just because it is OR. ?
It really is 'The Computer says No .....' mentality.

Quote
The real problem is that the speed displayed in the browser, after running different flash-based applets, suggest something impossible.
Yes

Quote
If we do a Sherlock - and eliminate the impossible - then it suggests we are being given incorrect data in one or more place.
Yes

Quote
First would be to ensure I was connected as expected (eg not using the neighbour's WiFi accidentally, etc) with plain ethernet cable, and wireless totally disabled.

I am using a gigabit ethernet connection via an ASUS Router to the modem (also gigabit) from a PC. (I am well aware not to use WiFi.  ;D)
I do understand TCP/IP networking and it is also impossible to 'accidentlly' use my WiFi or connect to any neighbours WiFi.

Quote
Second would be to confirm the full sync data from the modem - ie plain text from a telnet session, or by running the snapshot graphs, and copying out of the Plink file.

As HG612 Stats gets its stats via a telnet session, do not see the difference.
Unless you are implying that HG612 Stats is changing the data (Misreading/Corrupting).
Not very likely as it would have been noticed by now, I think.

Quote
Third would be to run a download as raw as possible, and time it manually. I would run FTP or WGET from the command line under Linux (or Cygwin under Windows), but I'm sure there are plenty of other options.

Can do. Don't doubt it will agree with the IP Profile (approx)

My point in posting the OP was to confirm that the tools PN are using and are totally reliant on are not working.
The PN Support staff are unable to question anything that comes from OR or their tools.
This makes solving issues almost impossible as there is a readymade 'brickwall' that we all bang our heads against.
I do not believe it is 'impossible' to talk to OR and solve issues
AAISP can manage it, it just comes down to a will to try.
Plusnet should have an advantage, being able to use BT 'back channels' to communicate with BTW/OR.
Yes, I know the 'chinese walls' that are supposed to exists, but 'Back Channels' are not official, are they ?

Title: Re: Aardvarks line deterioration
Post by: kitz on April 16, 2015, 08:41:08 PM

I wouldn't mind seeing a TBB speedtest though to look at httpx1 v httpx6.

Your upstream throughput seems low - is it always around 15Mbps?   I'd expect that to be somewhere in the region of 18.5Mbps for a 20Mbps sync.
Also can you provide me with a tracert to the BBC.   Ive noticed a couple of those performance tests show very high ping rates - but that isnt conclusive as it is known to often be out when it comes to ping time.


Just in case you missed my post above.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 16, 2015, 11:34:42 PM

I wouldn't mind seeing a TBB speedtest though to look at httpx1 v httpx6.

Your upstream throughput seems low - is it always around 15Mbps?   I'd expect that to be somewhere in the region of 18.5Mbps for a 20Mbps sync.
Also can you provide me with a tracert to the BBC.   Ive noticed a couple of those performance tests show very high ping rates - but that isnt conclusive as it is known to often be out when it comes to ping time.


Just in case you missed my post above.

Yes I did.

I have swapped to the HG612 to see if DLM is not setting G.INP because I am using a 'non-kosher' modem.
Cannot trim screencaps, PC running very Busy ...... keyboard very sloooow.

TBB Test
http://www.thinkbroadband.com/speedtest/results.html?id=1429221754554326155 (http://www.thinkbroadband.com/speedtest/results.html?id=1429221754554326155)

Previous TBB on Zyxel
http://www.thinkbroadband.com/speedtest/results.html?id=1427168235691925955 (http://www.thinkbroadband.com/speedtest/results.html?id=1427168235691925955)

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-16%2023.04.04_zps3ploqaxj.png)

(http://i1072.photobucket.com/albums/w364/GrapheneMan/Screenshot%202015-04-16%2023.03.55_zpskbxtriz2.png)

UPDATE:
Fixed truncated link

Title: Re: Aardvarks line deterioration
Post by: Chrysalis on April 17, 2015, 10:38:54 AM
first speedtest url doesnt work.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 17, 2015, 12:17:33 PM
first speedtest url doesnt work.

Thanks, fixed truncated 1st link.
Title: Re: Aardvarks line deterioration
Post by: kitz on April 17, 2015, 12:22:50 PM
Quote from: kitz
I wouldn't mind seeing a TBB speedtest though to look at httpx1 v httpx6.

The httpx1 v httpx6 test I wanted to see is this one (http://www.thinkbroadband.com/speedtest/speedtest2.html). 
Sorry, I forgot that it only shows up by default if you are logged in, if you're not then you have to click the "alternative speedtester".

Can you let us see a http1 v http6 test please
You can input the results direct into this forum using the code selection "BBCode (for forums; not supported on thinkbroadband forums) Displaying Graphs"

Quote from: kitz
can you provide me with a tracert to the BBC. Ive noticed a couple of those performance tests show very high ping rates

At 239ms your latency is looking very high on that TBB speedtest and there's a lot of jitter going on. 
Can you let us have a tracert to BBC please.

Burst activity of 82Mbps and an average of 71.7Mbps when you were syncing at 69670 should not be possible

(http://www.thinkbroadband.com/speedtest/button/1427168235691925955.png) (http://www.thinkbroadband.com/speedtest/results.html?id=1427168235691925955)

Quote from: kitz
Your upstream throughput seems low - is it always around 15Mbps?

I can see from the speedtest history that it is often much lower than it should be.  It also appears to have some correlation with the downstream being higher than it should.  :hmm:

Do you have a software firewall or Antivirus prog that could be interfering with speedtests.  Its not unknown for some AVs to cause strange speedtest results.

Quote from: aardvark
PC running very Busy ...... keyboard very sloooow.
Was it running busy when doing the speedtests, that could cause buffering and explain the jitter, and possibly some of the latency.
Can you perform the tests when the PC has nothing else running.


Quote from: aardvark
I have swapped to the HG612 to see if DLM is not setting G.INP because I am using a 'non-kosher' modem.

I wasnt going to get you to swap until I'd seen the http1 v http6 tests and I also wanted you to get an adsl info --stats immediately before you took it off, and then one from the HG612 immediately after it synced to compare.

However from the information on MDWS stats I can see that as anticipated the HG612 is giving you a lower sync.  SNRm is a tiny bit more bumpy but nothing to worry about as your error rates haven't increased.

Despite the HG612 being on overnight, you still appear to be g.inpless [i made up a new word!]
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 17, 2015, 01:09:35 PM
Quote from: kitz
I wouldn't mind seeing a TBB speedtest though to look at httpx1 v httpx6.

Can you let us see a http1 v http6 test please
Sorry, was not logged in. Test Done.
(http://www.thinkbroadband.com/speedtest/button/142927437509686630812.png) (http://www.thinkbroadband.com/speedtest/results.html?id=142927437509686630812)

Can you let us have a tracert to BBC please.
Quote
1st Attempt
===========

Tracing route to www.bbc.net.uk [212.58.244.70]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  192.168.1.250
  2    15 ms    14 ms    14 ms  lo0.12.central12.pcl-bng02.plus.net [195.166.130.153]
  3    17 ms    17 ms    14 ms  irb.12.pcl-cr02.plus.net [84.93.249.114]
  4    15 ms    14 ms    17 ms  ae2.pcl-cr01.plus.net [195.166.129.6]
  5    15 ms    14 ms    14 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  6    15 ms    14 ms    14 ms  kingston-gw.thdo.bbc.co.uk [212.58.239.6]
  7     *        *        *     Request timed out.
  8     *        *        *     Request timed out.
  9    20 ms    15 ms    15 ms  ae0.er01.telhc.bbc.co.uk [132.185.254.109]
 10    15 ms    15 ms    16 ms  132.185.255.149
 11    16 ms    15 ms    15 ms  bbc-vip115.telhc.bbc.co.uk [212.58.244.70]

Trace complete.


2nd Attempt
===========

Tracing route to www.bbc.net.uk [212.58.246.93]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  192.168.1.250
  2    14 ms    14 ms    14 ms  lo0.12.central12.pcl-bng02.plus.net [195.166.130.153]
  3    14 ms    14 ms    15 ms  irb.12.pcl-cr01.plus.net [84.93.249.113]
  4    15 ms    21 ms    14 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  5    16 ms    16 ms    15 ms  kingston-gw.thdo.bbc.co.uk [212.58.239.6]
  6     *        *        *     Request timed out.
  7    16 ms    16 ms    16 ms  ae0.er01.cwwtf.bbc.co.uk [132.185.254.93]
  8    16 ms    16 ms    16 ms  132.185.255.165
  9    16 ms    16 ms    16 ms  bbc-vip014.cwwtf.bbc.co.uk [212.58.246.93]

Trace complete.

Burst activity of 82Mbps and an average of 71.7Mbps when you were syncing at 69670 should not be possible

Yes, as I keep saying  ;D. I did put it down to the fact on-line speedtests are prone to be inaccurate. [ :understatement:  ;D]
(http://www.thinkbroadband.com/speedtest/button/1427168235691925955.png) (http://www.thinkbroadband.com/speedtest/results.html?id=1427168235691925955)

Quote from: kitz
Your upstream throughput seems low - is it always around 15Mbps?

I can see from the speedtest history that it is often much lower than it should be.  It also appears to have some correlation with the downstream being higher than it should.  :hmm:
Don't know as I do not regularly do Speedtests. I do not need to upload a lot, so do not notice the 'slow' speed.  :shrug2:

Do you have a software firewall or Antivirus prog that could be interfering with speedtests.  Its not unknown for some AVs to cause strange speedtest results.

Quote from: aardvark
PC running very Busy ...... keyboard very sloooow.
Was it running busy when doing the speedtests, that could cause buffering and explain the jitter, and possibly some of the latency.
Can you perform the tests when the PC has nothing else running.
The busy was just as I started to post the reply. (As usually happens to slow things down even more).
I am using Avast (latest version). Do not know if it interferes.
Not aware of any reported problems with Avast. Will stop Avast and unload Firefox & use Chrome just for Speedtests.


Quote from: aardvark
I have swapped to the HG612 to see if DLM is not setting G.INP because I am using a 'non-kosher' (read NOT BT approved) modem.

I wasnt going to get you to swap until I'd seen the http1 v http6 tests and I also wanted you to get an adsl info --stats immediately before you took it off, and then one from the HG612 immediately after it synced to compare.

However from the information on MDWS stats I can see that as anticipated the HG612 is giving you a lower sync.  SNRm is a tiny bit more bumpy but nothing to worry about as your error rates haven't increased.

Despite the HG612 being on overnight, you still appear to be g.inpless [i made up a new word!]
Sorry, I have nothing to lose trying the HG612 now and I would not put it beyond OR to now be getting picky over modems with the current issues re:G.INP etc  ;D
I am sure DLM does not like me ....... it sure has had a good play with my line in the past.  :( :no: ;D

Title: Re: Aardvarks line deterioration
Post by: Al1264 on April 17, 2015, 01:40:38 PM
Will stop Avast
I find that you just need to temporarily disable shields with Avast: Right-click the Avast icon in the system tray (assuming you have it running there), select 'Avast shields control' then 'Disable for 10 minutes' (longer times available if required).
Avast makes a little difference to my TBB speedtests, disabliing shields shows a much flatter graph, particularly on the HTTPx6.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 17, 2015, 02:10:53 PM
Will stop Avast
I find that you just need to temporarily disable shields with Avast: Right-click the Avast icon in the system tray (assuming you have it running there), select 'Avast shields control' then 'Disable for 10 minutes' (longer times available if required).
Avast makes a little difference to my TBB speedtests, disabliing shields shows a much flatter graph, particularly on the HTTPx6.
Thanks, that is what I meant  ;D
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 17, 2015, 02:19:51 PM
Kitz,

Four consecutive TBB Speedtests FYI & delight, Thanks  ;D

http://www.thinkbroadband.com/speedtest/results.html?id=142927923511495643264
(http://www.thinkbroadband.com/speedtest/button/142927923511495643264.png) (http://www.thinkbroadband.com/speedtest/results.html?id=142927923511495643264)
http://www.thinkbroadband.com/speedtest/results.html?id=142927932118773591074
(http://www.thinkbroadband.com/speedtest/button/142927932118773591074.png) (http://www.thinkbroadband.com/speedtest/results.html?id=142927932118773591074)
http://www.thinkbroadband.com/speedtest/results.html?id=142927941513863040392
(http://www.thinkbroadband.com/speedtest/button/142927941513863040392.png) (http://www.thinkbroadband.com/speedtest/results.html?id=142927941513863040392)
http://www.thinkbroadband.com/speedtest/results.html?id=142927946603782210323
(http://www.thinkbroadband.com/speedtest/button/142927946603782210323.png) (http://www.thinkbroadband.com/speedtest/results.html?id=142927946603782210323)

I would like to run these again when it gets 'Busy' tonight & early in the morning as well.

As you can see I have lost circa 10M from my original speed, courtesy of OR & DCoE plus a little crosstalk.  :(  ;D

Title: Re: Aardvarks line deterioration
Post by: kitz on April 17, 2015, 08:36:10 PM
Ok thanks for the tracert and the graphs.

I had been concerned at some of the high ping times recorded on some of the speedtests, so I wanted to make sure that you didnt have some sort of hidden interleaving profile.   Everything looks fine with your tracert.

The last set of TBB with Avast turned off look much more like they should.  Your upstream can be a bit slow to get going sometimes, dunno why that is, but its not anything Im too concerned about atm.   The main thing is that your downstream now looks correct.

It would be interesting to do a BT performance test with Avast turned off.
Title: Re: Aardvarks line deterioration
Post by: kitz on April 19, 2015, 10:35:10 AM
Aardvark:    I happen to notice that your SNRm and max attainable has started to slide over the past few days since putting the HG612 on. 
You are now at >2Mbps below your sync speed.

If the DLM was going to notice the HG612 and re g.inp you then I would have expected it to have done so before today.   
Its obvious that the Zyxel VMG8324 performs better on your line than the HG612 so its up to you whether you want to put it back on again.

I can understand if you want to leave things be, because it will show the slow deterioration.  It may muddy the waters if PN see you put the VMG8324 back on and they think the line has improved. 

Title: Re: Aardvarks line deterioration
Post by: kitz on April 19, 2015, 10:37:38 AM
PS
Did you ever get around to trying the BTw Performance test with Avast turned off and have things returned to normal?

PPS

Ive split this to its own topic.
Title: Re: Aardvarks line deterioration
Post by: AArdvark on April 25, 2015, 03:53:57 AM
I put the Zyxel back.
The line did not improve much and started to drift down to the same as the HG612 ??? approx 68xxx.
I noticed that the line would improve the max attainable by approx 1M and stay there for hours.
I re-synced at one of these points and have been synced at 697xx for 24hrs+.

It would be nice if G.INP came back but as I have seen elsewhere on kitz et al this is another 'kettle of fish'  ;D

The BTW Speedtests are normal for now.
Not chasing this at the moment, watching the 'G.INP Show' as the final state of my line has not been reached.
(I hope!)