'High' and 'Low' on the BT checker

Looking at the DSL checker for my line it reports speeds attainable of 80 'high' to 58.6 'low' (for a clean line) and 68.6 to 42.1 for an impacted line.

FWIW I have 60 down, but I was wondering what impacts on a clean line to vary the speed? Naturally I have no idea if my line is clean or impacted but even if impacted shouldn't I get near on 68.6? or is 60 near on 68.6?!

Also what (at the foot of the table) are Bridge Tap, VRI and NTEFaceplate. All of which report "N" for no I guess.

Not moaning, just wondering

A clean line - in the perfect case will be from the cabinet to you without any joints. The system will know your approximate line length and will base the estimates on that. The factors which will then affect the potential speed on a clean line include: actual line length, gauge of copper used, external noise sources, other users in the same multi-pair cable and the proximity to you, the actual quality and impedance of the modem at your end and more. They will have a cumulative effect on the overall speed you can get.

Impacted: Again several things which will cause a line to be impacted; bridge taps, bad quality wiring at some point which could be in your premises, localised high energy noise/interferer ...

And finally - no you should not be getting near 68.6 even if your line is impacted - you could still get as low as the estimated 42.1

Further to what MHC says, do you have any stats for your current set up ....

And how is it all hooked up in your house .....

despite it having an N for bridge tap, these are the most likely cause for less than expected sync rate.

Like this from my router?
Done my best to format it! <- oh well, never mind...


Receive Direction Send Direction
Max. DSLAM throughput kbit/s 80000 20000
Min. DSLAM throughput kbit/s 128 128
Attainable throughput kbit/s 60115 20050
Current throughput kbit/s 60115 19797
Seamless rate adaptation off off

Latency fast fast
Impulse Noise Protection (INP) 0 0
G.INP off off

Signal-to-noise ratio dB 6 6
Bitswap on on
Line attenuation dB 17 20

Profile 17a
G.Vector off off

Carrier record A43 A43

Which router is this?

Which cabinet are you connected to, ECI or Huawei?

I note G.INP is off.

Hmm, no clues to be had there especially ...

What of the set up please ?

For 17 dB line the upstream with an attainable of just over 20 and actual just under is about right - maybe just a little low. the downstream is quite a bit lower that I would expect - it should be up in the high 60s to mid 70s and with a 6dB SNR, I would be thinking that there is a significant noise source affecting a block of downstream tones. Some plots from DSL stats may show something.

Router: FritzBox! 7490

Cab type: Does Infineon 11.8.6 help?

Exchange POLLOK served by Cabinet 9

What should I know about G.INP

No complaints about the speed or service here BTW, but it helps compare and contrast possible speeds at other locations.

Stats

https://imgur.com/a/yaBFv

.... But still no description of the set up at yours .....

In reply to a post by Zarjaz:

.... But still no description of the set up at yours .....

Sorry, what would you like to know? The FritzBox! is connected to the OR box which was installed when I got fibre. They gave me a modem and router which I swapped out for the all-in-one FritzBox! I don't have any landlines connected to the line (the Fritz acts as a DECT basestation). There are a couple of Macs on the network using Ethernet. Phones and stuff by WiFi as normal. I do have permanent VPNs with another FritzBox! or two in other parts of the UK. All on Zen. FTTC 80/20.

In reply to a post by Woolwich:

Exchange POLLOK served by Cabinet 9

What should I know about G.INP

Ok so you're connected to an ECI cabinet so no G.INP for you yet.

Perhaps the chipset in the Fritzbox isn't very compatible with the ECI so does the Fritzbox give you the option to use a standalone modem? i.e. does it have an ethernet WAN socket?

If you still have the white OpenReach modem and it's a Huawei HG612 then it can easily be unlocked to obtain very detailed stats. Using it for a few days with a stats program like DslStats can help diagnose all sorts of possible issues.

So if it�s an original engineer install one would hope there�s no bridge tap internally.

I note you don�t use the voice side, but it may be worth plugging a hard wired phone and listening, you may have a good old fashioned noisy line ?

In reply to a post by BatBoy:

Perhaps the chipset in the Fritzbox isn't very compatible with the ECI so does the Fritzbox give you the option to use a standalone modem? i.e. does it have an ethernet WAN socket?

Yes, I could do that...

In reply to a post by j0hn83:

If you still have the white OpenReach modem and it's a Huawei HG612 then it can easily be unlocked to obtain very detailed stats. Using it for a few days with a stats program like DslStats can help diagnose all sorts of possible issues.

Again, yes but... I'm moving soon and my original question was an attempt to compare the potential speeds at my new location with what I have now then work out what my 'real' speed could end up at. The BT checker suggest in 76.4 to 55.8 down on a clean line. My thinking was that's lower than I have now so I'll end up with around 50 to 55 or so.

I think it would be interesting to stick the OR modem on and look at its stats but I don't think I'll have time to do this. It will also, no doubt, mess with my current setup and VPN connections which run overnight backups.

On the other hand could the position of my Mac mini (next to the FritzBox!) and a JBOD full of spinning hard drives be causing any interference? I guess I could move the Fritz a little way away and see. Would I have to reboot to get a new connection?

So are you moving to Pollok or from Pollok?

In reply to a post by Zarjaz:

So if it�s an original engineer install one would hope there�s no bridge tap internally.

I note you don�t use the voice side, but it may be worth plugging a hard wired phone and listening, you may have a good old fashioned noisy line ?

Hopefully...

Of course I forgot about the hardwired phone in the kitchen which somehow doesn't ring. [UPDATE: it doesn't ring cos I took it away, theres an extension but no handset.] The incoming OR line comes all the way through the house to the 'master' where the FritzBox! is connected. There's a cable returning from the master back through the house to the kitchen phone. I seem to remember after the OR engineer fitted the new fibre master box the phone in the kitchen didn't ring and there was fiddling to be done.

So there could be odd internal wiring...

And noise on the line. Years ago (back in ADSL days, when I used the landline) I could hear the occasional noise, a buzzing which would get a bit louder and then fade away. Not quite but almost like a doppler effect. BT/OR came and looked but, naturally as it was an intermittent 'fault' couldn't do a lot. I had two phone lines at the line so he just swapped them over. But I'm sure I could still hear it at times. I forgot about it as I never use the landline nowadays. May its still there.

I appreciate everyone's advice. Other than move the Fritz to see if the Mac and hard drives are causing interference I don't think I'll have the time to try the other suggestions. But when I move...

In reply to a post by BatBoy:

So are you moving to Pollok or from Pollok?

Away from Pollok exchange.

So if your new cabinet is Huawei you should get better sync speeds

In reply to a post by BatBoy:

So if your new cabinet is Huawei you should get better sync speeds

Looks like the Huawei 288 /384HD FTTC cabinet shown on this page http://www.kitz.co.uk/adsl/fttc-cabinets.htm

In reply to a post by Woolwich:

Again, yes but... I'm moving soon and my original question was an attempt to compare the potential speeds at my new location with what I have now then work out what my 'real' speed could end up at. The BT checker suggest in 76.4 to 55.8 down on a clean line. My thinking was that's lower than I have now so I'll end up with around 50 to 55 or so.

Once the choice of "clean" or "impacted" has been made (and it is a choice you can make), then the range of speeds applies as follows:

- You can end up anywhere within the range, or above it, or below it.
- The range is calculated from the performance of "similar lines"
- 20% of subscribers on "similar lines" go faster than the top of the range
- 20% go slower than the bottom of the range
- 60% are between the two figures.
- If you are in the bottom 10%, then Openreach will send an engineer to investigate.
- Barring a real fault, the main driver of where *you* will be is crosstalk
- Crosstalk is caused by other subscribers; it is random; you cannot affect it

The clean/impacted decision - it isn't about how the line is actually behaving. It is, instead, about whether the line is trusted to behave well... and that "trust" is gained by sending an engineer to install things.

Note that clean/impacted never existed in the checker at first. It only came into being when BT introduced self-install. The thinking at the time was that engineer-installations got a "clean" speed quote, while self-installed lines got a lower speed quote. The aim was to stop BT engineers being called out to someone who didn't pay for a "proper" installation.

Nowadays things aren't so clear, as ISPs have muddled their product offerings around over time.

So ... how do you *choose* clean or impacted? Simple: pick an ISP that quotes you the "clean" range in preference to an ISP that quotes you the "impacted" range or an ISP that quotes the combined range (as examples, Plusnet do the first; TalkTalk do the second; AAISP do the third).

If you want the best potential speed (and not just a contract term to argue with the call centre over), choose an engineer installation.

In reply to a post by WWWombat:

Once the choice of "clean" or "impacted" has been made (and it is a choice you can make), then the range of speeds applies as follows:

...snipped...

Thank you for that, very interesting info.

I'm currently with Zen and happy so I want to stick with them. I did have a conversation about the new address which they checked but I can't remember what speed they quoted. Its going to be a new build, I'll opt for the engineer for sure.

I don't know if I should ask in a new thread but what impact could the builder provided internal wiring have? The 'master' socket is likely to be in the living room where I don't want the router. They may/may not move it to under the stair where I do want the router. So the router will either be plugged into a master or extension socket which relies on whatever phone cable they run through. Should I be worried/bothered? Should I insist the master socket is where I want it? But I assume they'd need OR to install the socket under the stair rather than in the living room (where there's an external wall and the phone line pops through). Will OR just move the socket (for a fee) and use whatever cable is installed? I can't imagine they'd run a new cable inside the partition walls.

[BTW, this is the same property I was checking a few months ago where fibre was unavailable due to capacity. That seems to be sorted now but there's no FTTP]