If your estate lines were installed within the last 8 years, it won't have been done with aluminium.

I was just about to say what Zarjaz said.

My gut feel says that you should be getting better speeds for that distance, but the first question is... is that a realistic estimate for distance?

We can't tell directly (though an Openreach engineer can get an estimate off the JDSU). However, we can get a feel for distance based on the attenuation figures. They will be slightly different through copper & aluminium, and it can vary based on the diameter of each cable. However, it is the attenuation that affects the drop-off in speed more directly than the distance.

My current line is about 450 metres, and gets 40/10 with a max attainable of 90/26 according to the modem or 85/24 according to the JDSU.

For comparison, here are my line stats, although they are generated with the Unix commands, so they look a little different, and have different axis values.

Your modem shows attenuation figures of 10dB for D1, 22.5dB for D2, and 33.8dB for D3.

My equivalent figures are 11.4dB for D1, 27.3dB for D2, and 41.4dB for D3. As attenuation is linear, it suggests that your line is 20% shorter than mine - so around 350 metres.

So - attenuation looks OK, and probably means you are on copper, without any obvious problem in that regard.

The next thing to look at is the noise. The SNRM values from the "pbParams" section show yours to be 8.6dB, 8.6dB and 8.4dB - where mine are all about 21dB. This is what you'd expect after the modem reports a much lower maximum attainable figure - that there is much less margin. However, your TX power levels are higher, suggesting that the signal level is a little higher at the cabinet, and with less attenuation, will be higher at the modem.

That suggests that the problem isn't the amount of signal - so must be noise.

The QLN (quiet line noise) graph demonstrates this. As I understand it, a value of around -140 dBm means a pretty quiet noise level. My line shows values of -130 to -140 in the ADSL range, and then -120 to -140 in the higher VDSL2 range.

In your graphs, they show about -130 in the ADSL range, then -100 to -120 in the higher range. This suggests a fair degree of extra noise.

The shape of the QLN graphs is, I think, significant too. The general shapes are pretty similar - suggesting that your additional noise is coming from throughout the spectrum pretty consistently. My guess is that this would be more likely to come from crosstalk from another VDSL2 system rather than being interference.

I recently moved, and had an earlier FTTC line. That one was 650 metres long, and could only sustain 36/10 on the 8c profile (due to errors), but jumped to 40/10 on the switch to 17a. The modem predicts a maximum of 59/16.

The line stats for the original line show a few differences. Because of the distance, I can understand that the bit loading stops before the end of the D3 frequency band. However, the bit loading from tone 1250 upwards is very similar.

The thing I don't understand is why the bit loading is so much higher up to tone 1000 in that first line. Very strange....

That line too shows noise values in the region of -120dBm to -140 dBm. The only difference is that it seems to get quieter, but at the same frequency point where the signal can no longer carry any information - presumably the interference (crosstalk) can't be carried the same distance either.

Against those 2 graphs, your noise levels are definitely looking to be the problem, and throughout the spectrum.

Hi WWWombat,

Just for reference, regarding your mention of bit-loading, these are my line stats:-

Line Stats

Due to attenuation/distance from the cabinet, my connection is unable to make use of the higher frequency tone band plans & you will see high bit-loading in the lower band plan.

Having seen quite a few connection stats now, it does appear there is a pattern of those connections with the higher attainable rates having bit-loading more evenly spread across all the band plans.

Connections with lower attainable rates (but still higher than 45Mb or so), although still being able to make some use of all the band plans, also have a higher concentration of bit-loading at the lower frequency band plan.


Paul.

That's probably what I'd expect. In fact, it would help from a crosstalk perspective if those short line (but capped) modems would actively choose to use the higher frequencies in preference to using the low ones at all.

Looking at your graphs afresh... the noise levels are all around -140 dBm, suggesting you aren't suffering any noise.

Other than that, the only things that stand out are
- In the SNR graph, that the levels for D2 seem lower relative to D1
- In SNR, that D1 seems to curve downward as it goes to tone 700 and above
- In SNR, that the area from tones 100-300 seems patchy.

The last one is hard to tell, because I think the vertical lines are drawn too thick - and hide detail (at least on the Unix version). There they are plotted as style 'boxes' of width 0.1, and I wonder if they are better plotted as a style of 'impulses' instead.

How do you generate them graphs?

In reply to a post by WWWombat:

The last one is hard to tell, because I think the vertical lines are drawn too thick - and hide detail (at least on the Unix version). There they are plotted as style 'boxes' of width 0.1, and I wonder if they are better plotted as a style of 'impulses' instead.

I can’t recall the code in the Linux script at the moment & I am not at home to check.

The Windows script uses:-

“set boxwidth 0.1 absolute”

That was changed from the default setting to “absolute” to stop any boxes expanding in width to touch adjacent boxes as they were far too thick.
A lot of subtle differences were originally being missed.

& it plots the data “with boxes fs solid”

I’ll experiment with impulses, but I have a vague recollection that 0.1 absolute generated the thinnest boxes.

Do you use Linux or Windows?

There is a thread on use of the Windows variant over here.

If you use Linux, then you use the same link as on the first post of that thread, and download "graph.sh". The script works pretty well, but depends on having ImageMagick and GnuPlot installed.

In reply to a post by Bald_Eagle1:

“set boxwidth 0.1 absolute”

The Linux one does the same too. Using impulses doesn't change a thing.

However, the main problem is that we're trying to draw a plot with 4096 values (and a range of 4250) onto an image width of 640 including the axes & margins. There is bound to be a loss of clarity.

When I fiddled to adjust the size (and going direct to PNG rather than via PBM), I get far better definition.

This gets considerably finer detail, but I haven't worked out the perfect width yet.

There is an issue with your line / setup. Problem is BT say 15meg sync is acceptable so unlikely to get someone out to look - also NGA has dedicated teams who are given 2 weeks training, a van and a screw driver and are let loose on these.
Only the overflow jobs go to the real engineers and so why you were left with low speeds.
At 362M copper from cab on provision (3rd sub) I got full speed with attainable of around 140Md/s 54Mu/s - checked it a few days ago with tester and still sync 40/10 but SNR has shot up to 22db (from 6) and attainable rate is now 98Md/s 38u/s. I know the line is perfect - this just crosstalk from other subs.
One thing to check is whether he installed it correctly in your house. Do you have more than one NTE5? did he fit a proper filter plate or did he simply unplug the adsl modem and put the OR modem in its place?
You would be surprised at some of the awful setups I come across after these guys - often worse than if they had just allowed a self install. But hey - low quality = high productivity = bosses get bonuses - got to keep these things in perspective.