Inconsistent speed- low SNR high attenuation
daylight ? 
I think the WiFi/Ethernet thing is a red herring, it's almost certainly a modem or line fault, once that SNR is at 5-9dB things will stabiise a awful lot
Martin
I think the WiFi/Ethernet thing is a red herring, it's almost certainly a modem or line fault, once that SNR is at 5-9dB things will stabiise a awful lot
Martin
Battery symbol 
Well spotted i will certainly do as you guys suggest and swap to another spare modem to see if this makes any differene. This may be a stupid question but is SNR greatly affeted by the router?
i will certainly do as you guys suggest and swap to another spare modem to see if this makes any differene. This may be a stupid question but is SNR greatly affeted by the router?
Quote from: Simonleeuk Yeah I appreciate postings speed tests via wifi not ideal, but it's really just to give an idea of the vast inconsistency of my connection. My connection is pretty rubbish anyway so Ethernet results pretty similar.
Simon,
Can we slow down please? There is a danger in confusing matters by changing bits before we clearly understand where we are, due to there being a lack of clarity in what's being said and done.
First the speed tests - where the speed tests performed on a PC connected by Ethernet to the router with all other connections (phones over wifi etc) shut down? Looking at those graphics again I realise that they are phone screenshots, which might imply that you ran the speed tests on a phone over wifi. It is important that this is clarified, as I had assumed that the speed tests were performed over Ethernet and therefore the slow speed is down to line errors.
If this assumption is flawed, there is a risk of the wrong path being followed.
What are your grounds for believing your connection is "pretty rubbish"? Frankly getting the synch rate that you are achieving is quite stunning - there might be consequential errors, but the synch rate is extraordinary for your line's attenuation! I'm on a 50dB attenuated line and I struggle to achieve 5.5Mbps @3dB consistently. You've got a much higher attenuation (which is bad) and a much better synch speed.
At the moment, I have no reason to suspect the modem / ADSL performance - the error counts are zero. However I see you've gone ahead and swapped modems. This is not ideal when one is trying to diagnose a complex issue.
Quote from: Simonleeuk current speed test. ok at the moment but no doubt will drop again in an hour or so. Don't know why it says Sky broadband for ISP, possibly because I'm using an old sky router?
2.3Mbps is not "OK" on your line. Something more like 4.5Mbps would be more nearer the mark. I note that you have only posted router stats, your BT and PN profiles are also required to assess what is happening here - please see http://community.plus.net/forum/index.php/topic,96155.0.html
Further more the report that you are on Sky is most certainly disconcerting - can you please check the login credentials on this other modem. Those speed test systems deduce the IPS from your IP address. This rather suggests that somehow you've connected to Sky because the router logon still has the Sky details. Not really sure that this should happen, but I have seen the likes before. It could even point to a crossed line.
Can you do a "tracert plus.net" in a cmd window (please ask if you don't understand) and this will clarify exactly how and to what you are connected.
Quote from: Simonleeuk This may be a stupid question but is SNR greatly affeted by the router?
If you don't know the answer, there is no such thing as a stupid question, especially when you wait for the answer before acting. You might also ask a question which no one else thought to ask / see (like why does its say Sky?) because others were looking at the information they were seeking!
SNR is not affected by the modem, rather some routers are more tolerant of low SNRM. Also some routers calculate SNRM slightly different and thus the reported figure could be a little different compared to another modem. Some modems (most usefully) report SNRM in 0.1dB increments, others in 0.5dB steps and least useful of all in 1dB steps.
Quote from: Simonleeuk Thanks Townman for your suggestions. Yeh looking at other threads Plusnet moderators on this forum seem to be really helpful so hopefully they can have a look into this for me. Anyway way of prompting them to join the thread?
Do you mean the moderators or the Customer Relations Team (CRT) members? Moderators and other members such as myself are customers just like you who give of their time to help others - we could be around at anytime. Members of the CRT - PlusNet staff - are only here during office hours Mon-Fri. They will step into a thread when it is clear that their assistance is required.
In respect of this thread, we are still trying to establish a clear profile of the issue. Until this is clear, it would be unproductive to get CRT involved and be unwise to instigate BTOR action.
What is required is to clearly confirm that the problem is or is not related to your line and not your router / modem / wifi / internal phone wiring. Until there is a stable view of the issue and the metrics, it is not possible to progress in a clear direction.
So please, with which ever router you want to work with, can you please get the set of data requested above in the speed issues thread, with BT Speed Test performed over an Ethernet connection. Note that this is the bench mark speed test required to progress a line fault.
Cheers,
Kevin
No problem. I'll post some more stuff for you to look at tomorrow. I'd leave your current modem/router on for a while in the morning to collect some stats.
You can run speedtests as often as you like no problem with that.
It's very improbable that you've actually connected to sky as it won't be configured in the exchange, apart from which they are LLU not using BTw's circuits.
The 3dB margin is nothing to worry about on a 21CN connection, bearing in mind you said it was a new estate with new cabling it's probably a pretty good line and that's what the exchange DLM thinks otherwise it wouldn't have set the Target SNRM down to 3dB.
I'll post a bit more shortly when I have a further read of your thread here and got a coffee
Hi Simon, I've had a good read through and hope I haven't missed anything. Sorry the coffee didn't get me back here quicker
Firstly, using and testing over wireless. Wireless is vulnerable to all sorts of interference issues from a variety of electric devices from Microwave ovens to Baby alarms, as well as neighbour's routers using the same or overlapping channels, which is why one should carry out tests over ethernet to see if the same problem continues. Wireless devices can also fail. I know sometimes it can be damned awkward, but you have to persist to identify the causes. When you have a moment http://www.plus.net/support/broadband/wireless_broadband/wireless_signal.shtml is a reasonable read to learn most of the basics.
A few other things worth mentioning at this point, you asked about all your computer equipment and everything located in the corner near the master socket. It's highly unlikely that this would cause a problem in normal circumstances, but if something has gone faulty it probably wouldn't make much difference even if it were further away. We'll come back to this later, it's not something that one needs to investigate in the early stages particularly if you haven't changed anything since before the problem started.
As Kevin Townman mentioned, you can't do anything about your attenuation (unless there's a fault!) other than moving. That said your two modem/routers are giving different attenuation values so it's a bit tricky taking the achieved speeds into account to be certain which may be telling the whole truth. You mentioned your TP-link is using it's original firmware. I'd certainly investigate if there is later firmware for your model and version, you should be able to update it offline, they may have a mechanism for doing it on-line, I don't know. My preference would be to do it off-line.
The speedtest result showing Sky. This is more likely because you logged Sky as your provider some time in the passed when you previously ran some tests? If so, just change it next time you run a test.
Ok, now you talked about a dropping connection. Was this just a perceived thing as a result of using your computer and finding you weren't connecting to web sites, or did you notice/look for changes in sync speed (not download speed) in the stats? If the former, it could easily have been issues to do with wireless. I know the initial stats suggested the possibility of some disconnects but that might just be for some obscure unrelated historical matter, maybe lightning for example - you may not even have visibly observed it, it needn't be that obvious. In any event the stats will start afresh next time it's used, so we'll get a better idea then at some point.
The fact that your Target noise margin is at 3dB and the excellent sync speed suggest you've not had any serious issue with dropped sync otherwise DLM would have changed that.
I'd leave the Netgear on for a good few hours today if it's holding the connection OK and post some stats if you can now and later. See if you can include an uptime figure, and see if you can find any more detailed error figures on another page perhaps.
I'll post this, have another read to see if I've missed anything else worth noting at this point and post again. HTH.
Firstly, using and testing over wireless. Wireless is vulnerable to all sorts of interference issues from a variety of electric devices from Microwave ovens to Baby alarms, as well as neighbour's routers using the same or overlapping channels, which is why one should carry out tests over ethernet to see if the same problem continues. Wireless devices can also fail. I know sometimes it can be damned awkward, but you have to persist to identify the causes. When you have a moment http://www.plus.net/support/broadband/wireless_broadband/wireless_signal.shtml is a reasonable read to learn most of the basics.
A few other things worth mentioning at this point, you asked about all your computer equipment and everything located in the corner near the master socket. It's highly unlikely that this would cause a problem in normal circumstances, but if something has gone faulty it probably wouldn't make much difference even if it were further away. We'll come back to this later, it's not something that one needs to investigate in the early stages particularly if you haven't changed anything since before the problem started.
As Kevin Townman mentioned, you can't do anything about your attenuation (unless there's a fault!) other than moving. That said your two modem/routers are giving different attenuation values so it's a bit tricky taking the achieved speeds into account to be certain which may be telling the whole truth. You mentioned your TP-link is using it's original firmware. I'd certainly investigate if there is later firmware for your model and version, you should be able to update it offline, they may have a mechanism for doing it on-line, I don't know. My preference would be to do it off-line.
The speedtest result showing Sky. This is more likely because you logged Sky as your provider some time in the passed when you previously ran some tests? If so, just change it next time you run a test.
Ok, now you talked about a dropping connection. Was this just a perceived thing as a result of using your computer and finding you weren't connecting to web sites, or did you notice/look for changes in sync speed (not download speed) in the stats? If the former, it could easily have been issues to do with wireless. I know the initial stats suggested the possibility of some disconnects but that might just be for some obscure unrelated historical matter, maybe lightning for example - you may not even have visibly observed it, it needn't be that obvious. In any event the stats will start afresh next time it's used, so we'll get a better idea then at some point.
The fact that your Target noise margin is at 3dB and the excellent sync speed suggest you've not had any serious issue with dropped sync otherwise DLM would have changed that.
I'd leave the Netgear on for a good few hours today if it's holding the connection OK and post some stats if you can now and later. See if you can include an uptime figure, and see if you can find any more detailed error figures on another page perhaps.
I'll post this, have another read to see if I've missed anything else worth noting at this point and post again. HTH.