I looked at line 3 for comparison, which is 3.066Mbps downstream, so much faster downstream, dreadful upstream, and much cleaner
Live sync rates:
#1: down 2898 kbps, up 529 kbps
#2: down 2807 kbps, up 553 kbps
#3: down 3066 kbps, up 396 kbps
#4: down 2564 kbps, up 522 kbps
Line 4 is running at a really, really low d/s SNRM at the moment, because I forced resynchs at opportune moments when the SNRM was well above the target. But I don’t like the number of CRCs I’m getting compared with the other lines, so I forced another resynch in the middle of the night to get it back on the target of 3dB d/s as opposed to being so very low at 2.4dB. That has reduced the downstream synch rate somewhat but no errors now.
Line 4 QLN is fairly rubbish too, compared to line 3. Here’s line 4 QLN:
This is line 3 QLN for comparison:
So much quieter.
All this extra noise - I wonder if line 4 is better at
detecting it, it’s not that there’s more noise out there, surely? - apart from crosstalk from different neighbours, how can the noise environment be the different? The two conductors are in the same place, (approximately) in the same environment and should be exposed to the same noise, but the QLN says otherwise. Wild speculation: if there is a new non-linear joint in line 4, detecting noise, then that would make sense because whatever the cause of the phenomenon is, it is new: it’s only been around for a week or so; something was done to the line and this noise then appeared, and I lost 300-500k d/s sync which is around 10-15%.
Is that a sensible suggestion?
There’s no way that I can get BT to sort it out as the line is nowhere near faulty; I used to be incredibly lucky with it with extremely low noise and low attenuation for such a long line, and now I’m not quite so lucky any more. If it were to develop a fault then perhaps the repair process might somehow magically kick the current badness out the same way as it was introduced but in reverse. AA is doing their bit, constantly measuring the sync rate, collecting data.