I'm impressed by how quickly you were up and running. I'm afraid your modem doesn't produce enough detail for proper analysis like a HG612 would.

I was tempted to see if I could pick up a second-hand HG612 on eBay. Is it likely to perform better than the Technicolor? I do have a router (ASUS DSL-N55U) that I could use with it. Is the modem likely to make much difference to performance?

I couldn't say if it would perform better than your existing Broadcom-based device but it would allow more analysis of your connection.

Thanks, I'll have to google the stats issue as I have no real idea what the ones I do get really mean and what extra ones would be useful.

I guess I could get an HG612 on eBay, use it to check my line and sell it again when I have finished, so it wouldn't cost a lot.

I notice that, on the ping monitor, mine has a lot more yellow than most but the green bit went down when I moved to FTTC.

Edited by filbert42 (Mon 07-Dec-15 10:02:18)

To be honest the whole data collection suite is built around the unlocked HG612 and people have a lot of skill on the analysis of the data, especially on the Kitz forum.

In reply to a post by filbert42:

As far as I can see, my exchange (Suckley - WMSUC) has ADSL Max but not 21CN.

Yes - that is indeed the case.

In reply to a post by filbert42:

What I don't understand is why the BT wholesale speed test is so variable and often so far from reality.

I have to admit, I don't find it very trustworthy for giving me an idea of real throughput speeds. I find speedtest.net to be more reliable for me, especially when I ask it to use a London server or the Vodafone one in Newbury.

The BTW tester tends to be left for two purposes
- To see what IP profile is being reported by BT (for which you needto see the "extra diagnostics" at the end
- To run tests that the ISP can see

In reply to a post by filbert42:

If I look at the BT broadband availability screen, I get results that say that "FTTC Range A (Clean)" speeds are from 12.8 to 19.6 Mbps and "Range B (Impacted)" speeds are from 5.3 to 14.4 Mbps.

I am at the bottom end of "Impacted" and the screen explains what it means by "Impacted" but doesn't tell me how I can find out whether my line is in that category or not. Or whether things can be "fixed".

The screen gives you a flavour of what it means, but even BT internal documents aren't tremendously clear as to what the two ranges mean, and when they should be applied.

However, the checker was upgraded to cater for self-install FTTC, a couple of years back, and that upgrade is what introduced two ranges. That's a clue as to why two ranges exist...

The BTW FTTC Handbook advises ISPs that, when offering a speed estimate to end-users, they should use the impacted range for self-install.

However, it doesn't explicitly tell ISPs when to use the clean range - we just have to assume they mean that it applies to engineer installations. The theory being that an engineer's tests will ensure that any wiring issues or copper line conditions are fixed during the installation, whereas they would be left in place in a self-install.

That, at least, gives us an idea of how an ISP should use the ranges when giving you an estimate before installation.

The document also explains the circumstances when Openreach will reject the fault (the 10th percentile I mentioned earlier), but that doesn't mention which range Openreach will choose to use.

The document also doesn't cope with complex issues, such as where you start out with an engineer install, but migrate to an ISP with a self-install. What threshold should the new ISP use? Or the case where a self-install happens, gets poor speeds, and a subsequent engineer visit fixes things. Should this now have a clean threshold?

All in all, it leaves things murky for fault fixing.

The best thing you have to go on is the speed estimate you were given during installation. But even that can be wrong.

As for being in the 0.05% there are times when one would prefer to be among the masses! I think I might have to wait for Fibre To The Pole (which is just across the road from me) - which is what I think G.Fast means.

It is indeed frustrating to fall outside the norm sometimes.

Unfortunately, while G.fast was originally designed to work from the DP on the pole (hence the term FTTdp), it looks like it will work well at much longer distances, so BT are re-evaluating just where in the network they will site the nodes.

It seems likely that the first round of installation (2017-2020) is very much likely to deal with the area immediately surrounding the existing cabinets.

But who knows where the new iteration of on-demand fibre could go over the next few years?

The current position is that my ISP has asked BT to look into it and will absord the charge (I won't post the name before the whole process is finished, I don't think it would be fair to them)

What was your estimated speed when you signed up? And was it a self-install, or engineer-install?

In reply to a post by filbert42:

2.1 Type VDSL
2.2 Status Up
2.3 Uptime 1 days 21 hours 42 mins 7 secs
2.4 Line Rate (Down / Up) 6733 kbps / 1285 kbps
2.5 Noise Margin (Down / Up) 6.3 dB / 6.9 dB
2.6 Attenuation (Down / Up) 35.6 dB / 17.3 dB
2.7 Power (Down / Up) 9.6 dBm / 10.5 dBm

That downstream power level doesn't look very high.

I've had 3 different lines, and they've been:
+ 12.3 dBm at 650m
+ 13.1 dBm at 375m
+ 13.1 dBm at 125m

I don't think there is anything you can do about it, but I just thought I'd note it.

In reply to a post by filbert42:

Well, it finally came to our village. I found out by accident and was on the phone to the ISP the next day. The line tester was predicting speeds of 12-19Mbs. I couldn't wait...
Two weeks later, it was all done!!! YAY.

With bated breath, I logged in a did a speed test. Result before, (with ADSL) ~5Mbs, after (VDSL) ~ 6.2Mbs - wow!!!!

What an anticlimax. I have been on to tech support and we have tried various things but that is what it is. And, yes, I am plugged into the test socket.

The odd thing is that the BT wholesale speed checker tends to give results around 10Mbs (sometimes it's as low as 6 but 4 out of 5 times it's 10 or even a touch more).

I have tested with TBB speed test and Ookla, same result. If I download one of the TBB test files, sure enough, around 6.

I'm about 1500m (by road) from the cabinet (which, ironically is a about 100m the *other* side of the exchange).

The ISP says that they can't log a fault with BT as it is within the "tolerance band" or some such excuse.

So, is there any point to FTTC, is it all a just a big sham???

By the way, if I try to run the 'further tests' on the BT speed-checker, it fails with the message 'The Performance Tester is currently unable to run a speed test for your broadband connection.'

The problem is that to "promote competition" Ofcom insisted BT allow other companies to install LLU kit in exchanges, meaning that each exchange contains 3-5 sets of the same equipment for different providers. This means they have to keep old non fibre ADSL running (rather than switching everyone to VDSL or even ADSL from the cabinet) and they have to hamstring VDSL (FTTC) in order to avoid any risk of impacting ADSL.

And the benefit of this? We can buy the exact same product from lots of different providers who all increase their prices at once every year by the same amount like a cartel. Yay, or something.

A quick update - my ISP sent a BT guy out earlier this week and he confirmed that there are no specific problems on my line - it is just too far from the cabinet for anything better.

Apparently, we (me and the ISP) have 90 days to cancel but only if performance drops below 4.5 Mbs. BT seem to consider the performance acceptable.

Zen (the ISP) have been very good. They have arranged a call from BT which they have had to pay for as BT don't accept that there is problem. They have also been very good at responding to my emails and calls.

This is a very old thread now but I thought I ought to be fair to Openreach and update to say that they have been doing some work in this area (I have been told that they've been replacing aluminium cables but I have no direct evidence).

Anyway, whatever they have done, my typical speeds have increased significantly. In the middle of last year, they went up to around 13mps. Earlier this year it was 16 and the last month or so it's been a steady 17.5mbs.

That's plenty for what I mostly use.