Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Black Sheep]

Ah right. Cheers.

[Bald_Eagle1]

I just noticed this, is it expected that the interleaving would move up and down?

With such a sudden change in your SNRM values, I just has a feeling that Interleaving depth would change.
Mine has moved around quite a bit in the past (see attached).

Also what dya mean the cabinets dslam has not been updated?

Last October, we noticed a change in our band plan tones.

A few days later, our modem firmware was updated, resulting in another slight change.

Putting two & two together, we deduced that the DSLAMs' firmware was firstly updated in preparation for our modem firmware updates.

What I was getting at was that as your updated modem's firmware isn't necessarily matched with the band plans delivered by the 8a profile DSLAMs, it may possibly have a negative effect.

TBH, we know very little about the 8a profile as we started on 8c & were then switched to 17a.

[bilbokitz]

With such a sudden change in your SNRM values, I just has a feeling that Interleaving depth would change.
Mine has moved around quite a bit in the past (see attached).

Yes, I am a bit confused about this as I would have though it would change aswell + QLN got worse, still hasn't and interleaving is set according to line conditions here too.

I also find the interleaving level at work confusing given the slower connection longer line and lower noise margins ? Is this qln from work stats better than mine, would that explain it?

[bilbokitz]

More changes today, looks like I lost sync early on this morning, heres the resync. Attainable down again, snrm down up slightly and ds INP a bit better. The QLN graph is 'fluffier' but better I think?

[Bald_Eagle1]

With such a sudden change in your SNRM values, I just has a feeling that Interleaving depth would change.
Mine has moved around quite a bit in the past (see attached).

Yes, I am a bit confused about this as I would have though it would change aswell + QLN got worse, still hasn't and interleaving is set according to line conditions here too.

I also find the interleaving level at work confusing given the slower connection longer line and lower noise margins ? Is this qln from work stats better than mine, would that explain it?

It looks a little better to me i.e. more of it around or below the -140dB mark. especially when compared aginst your previous QLN levels (example attached).

[Bald_Eagle1]

I have extracted & attached just the QLN data from your montage to make comparing it against your previous QLN graphs a little easier (using a like for like format).

BTW, the dead flat lines in your work QLN graph is at -160dB. i.e. no reported US data.
That's not unusual though.

My graphing format intentionally doesn't show -160dB QLN data.

The attached graph does look quieter than the comparison graph, but the 'fluffiness' does suggest some sort of interference.
As QLN data is only obtained during a syns/resync process, it's hard to say if the 'interferenc' is constant.

If you are still recording ongoing data, is there anything also showing in the SNRM graphs?

[bilbokitz]

I had disconnected the laptop so had not been collecting stats at the time of the change. It looks like that was just a wobble, I have reverted to the older firmware with unlocked gui as the internet 'feels' better. Maybe as you say the newer fw is refined for bt 17a.

Everything has been stable over the last few days. Interleaving levels the same and I have been collecting the stats.

What figures provide the attenuation rate and therefore approx line length from the below, average of D1 & D2 line?

  VDSL Band Status        U0      U1      U2      U3      D1      D2      D3
  Line Attenuation(dB):  3.9     17.9     N/A     N/A    9.7     22.0     N/A
Signal Attenuation(dB):  3.9     16.9     N/A     N/A    9.7     22.0     N/A

[Bald_Eagle1]

We have never found an answer to your question & have been unable to work it out.

It's not just an avearge though.

It may possibly be the attenuation at a given frequency, apparently similar to ADSLs attenuation value.

If you have a look at your Hlog graphs & raw data, you may be able to 'guess' at an overall attenuation value, probably around the value listed under D1 (somewhere around tone 500 in the Hlog graph).

[bilbokitz]

Here's my last 14 days of stats.

Nothing much as changed as far as I can tell, and the extra noise that appeared never went away so guessing you are correct and it is indeed crosstalk from new users.

Do the FEC and CRC errors look okay, from readind around they dont seem too bad to me.

[Bald_Eagle1]

It all looks quite steady to me.

The large drop in attainable rate won't affect your restricted sync & throuput speeds, so nothing lost really (unless they do roll out 80/20 in the IoM).

As it was such a sudden drop in SNRM, I assume it was all down to a single user being added.
Mine went down in small, but clear steps over a long time period, always returning back up to 6dB at a resync with a loss of speed.

It happened again, 6 days ago.

DS Sync speed droped again, from 21987 Kbps to 18937 Kbps 

[bilbokitz]

Hopefully you will get vectoring soon. Apparently vectoring and bonding is being tested within the main exchange here, although "very much a test phase". The issue there will be that the main provider rolled put dgnd3700s to begin with which now don't even officially support vdsl. Its the fritz 7390 being rolled out now.

How far are you from the cabinet? I am 350 m by the most direct route but reckon 400 to 500 if I follow the estate and floor chambers etc.

[Bald_Eagle1]

I can't really justify the likely additional cost for bonding, but if Vectoring gets me back up to around 30Mbps/7Mbps as it was before crosstalk decimated it, I'll be reasonably happy.

My line length is between 1000m & 1100m to the NTE5 socket (engineer confirmed route) allowing for 'slack' & loops etc.

[bilbokitz]

Yet your interleving level is lower than mine, not sure I get why that is except to assume although a shorter line, its noisier.

Actually your qln is better.

[bilbokitz]

Hi Kittens, theres been some developments here in the Isle of Man, apparently VDSL "Plus" is being rolled out in feb which will be 80/10 up from my current 40/2.

I checked today looks like the incumbent provider has upgraded profiles to 17a!!!  though I am still capped at 40/2.  Attainables have shot up, I guess the profile make a difference to attainable.

[Bald_Eagle1]

You now have more tones available for bitloading using the 17a profile.

It seems you are now using the amended 'Blob' modem firmware version from a long time ago though (unless it was auto-updated).

Assuming the IoM uses the same DSLAMS & firmware as on the mainland, you may see some benefit from using the latest available DS/US tones via HowlingWolf's firmware version & the BT remotely updated software version:-

xdslcmd --version
xdslcmd version 1.0
DSL PHY: AnnexA version - A2pv6C038m.d24j
******* Pass *********


equipcmd swversion display
software version: V100R001C01B030SP08
xdsl firmware version: A2pv6C038m.d24j
cpu version: BCM6368
cfe version: 1.0.37-102.6