Print

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

[konrado5]

I've written already about this, however, I'm not sure if I explained it clearly earlier.
I've noticed three types of downstream SNR margin fluctuations. I think on the first two graphs I see normal SNR margin fluctuations but on last graphs I suspect either short lack of REIN or HR fault (very little of course).

I've noticed also erroneous upstream SNR margin measurement on the new synchronization until some bitswaps happens. On some following graphs there is noticable upstream SNR margin nearly 0 dB. It appeared when I did reconnect and got nearly identical upstream synchronization rate.  I've seen it many times. I've seen also many times SNR margin about 4.0 dB during first 5 minutes of the new synchronization and after 6.6 dB (but 6.6 dB after bitswaps). If I check SNR margin immediately after synchronization establishment I see always 6.0 dB, however it is often 0.2 dB (sometimes about 3-4 dB) after few seconds. What it happens? I'm almost sure it is erronously measurement. What is cause of it?

Best regards
konrado5

[kitz]

Why are those graphs so big.  I note the img says 1920 x1000, which means a normal PC screen is going to struggle to see them, nvm anyone viewing on a mobile device. 

and... from what I can see..... all graphs look perfectly healthy.  Im sure youve been told before that such small deviations are perfectly normal for a line your length.  In fact its very good... particularly since you are running at such a low SNRMargin, which youve been told countless times is not recommended.   Im surprised your line is holding on. 

>>> I suspect either short lack of REIN or HR fault (very little of course).

NO .
Millions would literally kill for your line.  Ive shown you before what a HR fault looks like.. it looks nothing like that.
Every single line except the very shortest would expect to see some random fluctuations of up to 1dB...  its normal... its just the way ADSL works.

[konrado5]

Every single line except the very shortest would expect to see some random fluctuations of up to 1dB... 

But on two last graphs there are momentary raisements about 0.6 dB. I say about very little HR fault

[kitz]

Konrado you are worrying over nothing on this one.

Read again up to 1dB that includes 0.6dB.... the long a line then the more the fluctuation.  Some long lines fluctuate by up to 6dB on a daily basis..  some even up to 12dB.. thats precisely why Target SNRMargins are required.

Youre running your line without any Target... which isnt healthy and Ive no idea why you continue to do so.   Youve been repeatedly told its not a good idea.
Even on a very short line that hardly fluctuated and flatlined I still would not run my line at less than 3dB...  that Margin is needed for a reason.   Even on a 7dB atten line I still occasionally saw the odd fluctuation of 0.5dB.  Its normal..  in fact for your line length, its bleeding fantastic.

That is not HR fault.. its symptomatic of tiny fluctuations from normal background noise.

Im not telling you again.  Ive nothing more to say.

[konrado5]

Read again up to 1dB that includes 0.6dB.... the long a line then the more the fluctuation.  Some long lines fluctuate by up to 6dB on a daily basis..  some even up to 12dB.. thats precisely why Target SNRMargins are required.

I have trouble with translating it. I think about fact that 0.6 dB higher SNR margin is only during a very short time (for example minute or 30 seconds).

Its normal..  in fact for your line length, its bleeding fantastic.

My line is under the ground. How long is upstream SNR margin the same on 7 dB attenuation line?

Best regards
konrado5

[kitz]

>>> I think about fact that 0.6 dB higher SNR margin is only during a very short time (for example minute or 30 seconds).

Exactly - its normal.

Lets have a look at my own very short line - also underground.  The adsl 2+ attenuation was 7dB.  I can look out of the window and see the exchange. We are talking a couple of hundred meters.

Usually its flat-lined.. but I still see periods with random bumps of 0.25dB... this is still classed as steady and normal. 

1) Flat-lined
2) Small 0.25dB fluctuations
3) little bit wavy


to be continued so I can add more screen caps

[attachment deleted by admin]

[burakkucat]

Random events will happen, especially with electronic equipment.

Having re-sized your graphs (re-attached, below), I can see those short duration increases in the downstream SNRM that concern you. However there is nothing that can be said . . . those increases are well within the experimental error range and, to me, do not signify anything at all. 

What would be of interest is if you were to set the target SNRM to 3 dB, gather 24 hours worth of statistics and then show all the resulting graphs.

[kitz]

...  continued


4) Blip 1
5) Blip 2
6) Closer look split bands.


See how on graphs 4 &5 I had a few blips - these are random and can be anything....  just the way adsl works from time to time - some background noise somewhere.   Theres no fault and nothing that can be done.

Now look closer in 6 which shows s breakdown, look how wavy my SNR is from time to time.

All of these graphs are completely normal and expected behaviour, theres nothing wrong with my line its just reacting to some background mush  type noise.   Theres nothing that can be done because adsl will always be susceptible to background noise.

Bear in mind my short line... so for a medium length line you get very good results, and better than Id expect for most medium lines.

[konrado5]

Thank you very much for reply. What do you think about it: I haven't got SNR margin fluctuations about 0.6 dB for upstream. There are blips only on downstream.

Moreover, I've written about other issue:

I've noticed also erroneous upstream SNR margin measurement on the new synchronization until some bitswaps happens. On some following graphs there is noticable upstream SNR margin nearly 0 dB. It appeared when I did reconnect and got nearly identical upstream synchronization rate.  I've seen it many times. I've seen also many times SNR margin about 4.0 dB during first 5 minutes of the new synchronization and after 6.6 dB (but 6.6 dB after bitswaps). If I check SNR margin immediately after synchronization establishment I see always 6.0 dB, however it is often 0.2 dB (sometimes about 3-4 dB) after few seconds. What it happens? I'm almost sure it is erronously measurement. What is cause of it?

Thank you very much for all your time
konrado5

[Loading]

  Could just be interfence affecting the downstream tones, you are just asking the same questions, after they have already been answered, there seems to be niothing wrong with your line, mine may jump around by 2.0db every couple of seconds, but there is nothing that can be done about it!

[konrado5]

Loading: as my line jumped around 3.0 dB every half of minute for upstream, my ISP did something and my upstream SNR margin jumps around only 0.2 dB.

[Loading]

Did you ever ask your isp what they did?

[burakkucat]

Thank you very much for reply. What do you think about it: I haven't got SNR margin fluctuations about 0.6 dB for upstream. There are blips only on downstream.

Random events will, statistically, occur at higher energies. The higher energies equate to the higher frequencies. The frequency spectrum that you are observing extends from approximately upper audio (kHz) to VHF radio (MHz). The energy increases linearly across that spectrum, directly proportional to the frequency. (Reference: Planc's constant.) Dividing up that spectrum into "named chunks" we have POTS (at low, audible frequencies), then US xDSL, then DS xDSL (at HF / VHF). The DS xDSL "chunk" of the spectrum carries higher energy than the US xDSL "chunk" and the latter carries more energy than the POTS "chunk".

It may seem odd but that's the way it is. I see nothing wrong. 

[Edited for spelling correction.]

[konrado5]

Thank you for reply. But it is still second issue: SNR margin 0.3 dB for upstream until bitswaps appear.

I've noticed also erroneous upstream SNR margin measurement on the new synchronization until some bitswaps happens. On some following graphs there is noticable upstream SNR margin nearly 0 dB. It appeared when I did reconnect and got nearly identical upstream synchronization rate.  I've seen it many times. I've seen also many times SNR margin about 4.0 dB during first 5 minutes of the new synchronization and after 6.6 dB (but 6.6 dB after bitswaps). If I check SNR margin immediately after synchronization establishment I see always 6.0 dB, however it is often 0.2 dB (sometimes about 3-4 dB) after few seconds. What it happens? I'm almost sure it is erronously measurement. What is cause of it?

[konrado5]

I'm not sure if I've written above clearly. It is hard to translate for me.