It's near a couple of extension sockets which includes a small mini fridge, however they're on 24/7 so it would stand to reason that it should occur at every other time too.
I did have the modem away from them before I bothered to screw it to the wall, and it made no difference then.
I guess it is something outside if its within a set time range, but random within there how bad it can be/when it happens. The 5/7pm spikes today were the earliest I've seen it.
Wish I could push for BT to hurry up with G.INP on the cab already.
Just as an experiment try moving the mini fridge away from the device if you can for a couple of days. Preferably place in on a different level in the house (IE if router is downstairs put fridge upstairs or vice versa).
It is possible that could be causing it (I frankly doubt it but its worth a try). Some of the cheap ones (with a tiny compressor and a Fan in the back but no cooling matrice) do cause electrical interference.
If router is plugged into an extension cable also try to eliminate that, again i doubt it will help but initially you want to try to eliminate anything obvious.
G.INP is unlikely to help much if at all on severe REIN, typically that helps more with general interference/line noise rather than REIN, its a tech based on tone allocation and if the REIN is bad enough to cause specific tones rather than the whole band to suffer there is not a lot G.INP is likely to be able to do. Would not matter how many bits got allocated to the REIN tones the REIN would still be there in pulses.
G.INP won't magically reduce the interference, I hope you're aware of that
What it will do is to significantly reduce the errors on the line (I had zero when I had it before DLM was reset) and thus sync speed will likely be higher and pings lower.
Vectoring will help with interference although it is primarily aimed at combatting crosstalk.
I'd be curious to see what the AC balance on your line is like. If you get another engineer ask them as the higher it is, the better the line is at rejecting interference. Above 50 is considered fine but ideally it would be 60+ (I've been told some pairs have AC balances of 80 or 90).
G.INP will not help bad or intermittent REIN its a tone allocation tech in simple terms on a normal line a persons tones are normally lower at specific points, G.INP identifies those and allocates bits to them, making for a smoother/more level tone allocation throughout the range and in most cases leading to a slight speed increase (typically 2-8Mb ish). Latency is increased when the tone allocation is taking place as obviously the protocol has to first think and figure out where to allocate things, which takes time.
Vectoring will not help at all, in fact vectoring on a line with REIN if anything could make it worse. The higher the tone frequency the more chance it will be affected by interference. Think of that like wifi and 2.4ghz vs 5ghz yes one can go faster but not as far and is more prone to errors. Vectoring when that (if that) gets here will basically for the most part affect short healthy lines more than long or noisy lines. The common denominator which will always stop long or poor quality lines improving on FTTC is the copper no matter what tech you employ you can not break the laws of physics or ohms law.
AC balence can only be measured on the analog (copper side) AFAIK (though i have not been in the job for more than a few years) of the line, measuring it at an end users premises which has FTTC is counter productive as you are not measuring the entire loop (though again maybe there is a way now). Voltage would be far more interesting and if that fluctuates.
Small correction NORMALLY only in the early hours, Ive seen it make alterations as late as 11.30am in the morning, typically it kicks in between midnight and midday. With minor changes (IE interleaving being enabled or disabled) occurring between 1-6am.
It happened at 6am because the ES went over the limit because of the huge burst of 1200 ES in a single hour, noticed interleaving is back on.
If its confirmed REIN there's really sod all I can do about it since it can't be internal.
If the interleaving helps (measure your line again over the next few days) then it is possible the REIN is not too bad and your line may go back to fastpath, the DLM system will likely have a fight with itself as to what it thinks will be best LOL. The bad side if the REIN is not that bad and interleaving helps it is that means the source will be even harder to find.
