In reply to a post by wiggers42:

That's the most sensible explanation I've heard.

Thats true, not much that Zarjaz doesn't know about this sort of stuff, its his bread and butter.

In reply to a post by Pheasant:

In reply to a post by wiggers42:

That's the most sensible explanation I've heard. Although surely once you reconnect the faceplate the system will revert to how it is now?

Quite.

That's the point. The demarcation / point of responsibility for Openreach is the master socket. Anything beyond that; I.e. any internal extension cabling, filters, other devices etc. are (a) completely your responsibility (b) may be impinging on the quality of the service, hence affecting your speeds.

Therefore Openreach only want to test the basic circuit with nothing else in the way. That's just logical, isn't it?

I wasn't aware of this demarcation but it makes sense now. I was on the chat with Vodafone for ages and the support droid was unable to explain any of this.

I've taken screenshots of the router status page showing line speeds before and after changing from faceplate to test sockets and I'll continue to over the next few days. Initially there was an improvement in upload speed.

In reply to a post by wiggers42:

I was on the chat with Vodafone for ages and the support droid was unable to explain any of this.

They live in the Potemkin village and if they maintain an illusion of industry, they are probably meeting expectations.

After leaving the router connected to the test socket all weekend the Line Status looks pretty much the same today as it did last week:

Line Quality
Downstream Upstream
Current Rate 66999 kbps 18312 kbps
Maximum Rate 77502 kbps 18320 kbps
Signal-to-Noise Ratio 8.3 dB 5.8 dB
Attenuation DS1 11.8 dB, DS2 28.3 dB, DS3 43.1 dB US0 3.2 dB, US1 21.1 dB, US2 32.7 dB
Power 13.4 dBm 6.9 dBm
CRC Errors in last 3089 minute(s) 1 2539
K (number of bytes in DMT frame) 0 0
R (number of bytes in RS code word) 8 14
S (RS code word size in DMT frame) 0.0754 0.4170
D (interleaver depth) 16 1
Delay 0 ms 0 ms

I'm sorry I don't know how to make this into a table. The downstream rate hasn't changed at all from 66999 but the upstream changed from 18011 (with faceplate fitted) to 18312 (without). The thinkbroadband speedtest at my PC (over mesh WiFi) is giving 62Mbps, which better than it was before I started complaining. Unfortunately I didn't note the line speeds at that point.

So there has been some improvement, look at theoretical max data rate, and the downstream SNR. Through put speed tests may be affected by sync cap which is not moving especially quickly.