Seriously high pings 600ms

[mattpearce79]

Just joined IDNET from an appauling service at Virgin ADSL...first of all the the consistency in my speed is wonderful compared to Virgin. My main issue is with my high pings 600ms on http://www.speedtest.net

I am using a wireless Netgear connection, i have downloaded pingplotter but every test tells me that "the destination is unreachable"

I have tried a tracert in the command prompt and the all time out - I'm lost and don't know what to do???

Anyone advise please?

Cheers

Matt

PS my pings were the same with Virgin but i put that down to their over congested network

[Rik]

Hi and welcome to the forum. Have a karma. 

What sites are you pinging, what happens if you use a cabled connection to the router? Ping Plotter is not always a good tool, try Ping graph from http://www.vwlowen.co.uk/internet/files.htm.

I just checked to www.bbc.co.uk, and got an average 24ms, with interleaving on, so it isn't a universal problem. What part of the country are you in?

[Inactive]

My result from speedtest;



Welcome to the forum and to IDNet.

Have a Karma.

[mattpearce79]

hi Rik thanks for the quick response...

Unfortunately using a cable connection isn't really an option for me. It's a logistical nightmare, I have no power supply near my phone socket and my router is situated in my old office (now my daughters bedroom) computer downstairs about 20ft away from socket. I do have a laptop so i could always try connecting via cable to that and see what happens.

If i get lower pings connecting via a cable, therefore establishing that it is my wireless connection that causing the high pings - is there anyway to improve it.

I am in yorkshire - on the South Elmsall exchange.

Cheers

Matt

[Gary]

My pings to  bbc.co.uk are 27ms so no issue here

[Inactive]

These may be an option;

http://www.ebuyer.com/product/109683

[Rik]

I do have a laptop so i could always try connecting via cable to that and see what happens.

Hi Matt

That would be a good start point, we need to eliminate the possibility that it's the wireless connection which is causing the issue. If a wired connection is the same then it's time to start digging, but if it cures the problem, then it's a matter of establishing what's affecting the wireless, eg another nearby network on the same channel.

As Inactive suggests, mains networking my be an alternative. Cheaper would be running ethernet cable, which can be up to 100m in a single segment.

[mattpearce79]

i'll try it later on tonight - i'm at work at the moment. thanks again

[Rik]

Let us know how it goes and we'll dig further from there.

[Lance]

It may be worth mentioning that the pings on speedtest.net, much like the speedtests, are highly inaccurate and next to useless. 

[Rik]

Or just completely useless...

[Gary]

I think completely useless is about right Rik, going by this

[Inactive]

Bah, it is that new router Gary..miracle thing.

[Gary]

Bah, it is that new router Gary..miracle thing.

oh indeed In, special batch you know 

[Inactive]

There would be a lot of used 2700's on the market ..if it did give those results Gary.. 

[Sebby]

My main issue is with my high pings 600ms on http://www.speedtest.net

I suspect this is down to SpeedTest.net, which is notoriously inaccurate. What happens if you ping www.idnet.co.uk or www.bbc.co.uk from a Windows command prompt?

[colirv]

This particular Ping Graph gives you the option to ignore the first ping time (to the router).

Edited to say that I'm now not sure what ignoring the first ping actually means - it could mean just ignore the first ping in the series.

[Rik]

As I understand it, Colin, it does indeed ignore the ping to the router.

[mattpearce79]

In reply to Sebby:-

"What happens if you ping www.idnet.co.uk or www.bbc.co.uk from a Windows command prompt?"

i think i did the trace correctly..i typed trace rt www.idnet.co.uk in th ecommand prompt and all the tests time out?!?!?

[Sebby]

No, that's something different (although it shouldn't be timing out, but we'll come to that later).

From a command prompt, type ping www.bbc.co.uk.

[mattpearce79]

"request timed out"


Packets sent 4
Packets lost 4

[mattpearce79]

i had a look at my router setup page and under the diagnostics section it allowed to to do a ping test:-

PING 212.58.224.131 (212.58.224.131): 56 data bytes
64 bytes from 212.58.224.131: icmp_seq=0 ttl=122 time=45.0 ms
64 bytes from 212.58.224.131: icmp_seq=1 ttl=122 time=45.0 ms
64 bytes from 212.58.224.131: icmp_seq=2 ttl=122 time=45.0 ms
64 bytes from 212.58.224.131: icmp_seq=3 ttl=122 time=45.0 ms

--- 212.58.224.131 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 45.0/45.0/45.0 ms

[Sebby]

Well, that's a significant improvement over 600ms. I get about 33ms on an interleaved line. Do you know if you're interleaved and how far from the exchange you are?

[mattpearce79]

don't know if i'm interleaved and i'm 2.99 kilometres from the exchange (as the crow flies).

[Sebby]

It's not terrible if you're interleaved. What router do you have?

[mattpearce79]

Netgear DG834PN router

[Sebby]

I'm not sure how you get the interleaving status on that router. Does anyone else know?

The ping of 600ms you got on SpeedTest.net is likely to be highly inaccurate. Other than saying 45ms is reasonable if interleaving is enabled, I can't really say a lot more. Perhaps someone else will be able to shed some more light.

[mattpearce79]

thanks Sebby

[Rik]

i typed trace rt www.idnet.co.uk in th ecommand prompt and all the tests time out?!?!?

Did you include the space, Matt. Tracert is one 'word'.

[Rik]

I'm not sure how you get the interleaving status on that router. Does anyone else know?

If it follows other Netgears:

Enter this in the browser address bar:

http://192.168.0.1/setup.cgi?todo=debug

which should result in a screen which just says 'Debug enable'.
(That's assuming that you haven't changed the router's IP address from the default).

Then exit from the web interface and open a command-line window. Type:

telnet 192.168.0.1

You should get a BusyBox welcome message to confirm that your telnet connection is established.

Now type:

cat proc/avalanche/avsar_modem_stats

[mattpearce79]

Did you include the space, Matt. Tracert is one 'word'.

I did type it as one word yes...i'll check the debug thingy when i get home tonight.

On an another note, before i joined IDNET i was with Virgin and was using their Netgear DG834g router that was locked and wouldn't allow you to update the firmware, so i've swapped back to my previous Netgear DG834PN.

Since swapping routers and joining IDNET my sync speed has dropped to 1400 kbps from 1856 kbps. When i first joind Virgin back in May last year i was syncing at 3000kbps, can i expect my speed to rise over the next 10 days? and if it doesn't is there anything i can do to increase it?

Thanks again for all your help so far!

Matt

[Rik]

Hi Matt

Has the sync been dropping steadily since the 3000 days, or did it only start when you moved to IDNet? If the former, it would suggest either changes in your internal wiring, poor filters or a line fault. If the latter, it would suggest a line problem.

How is your phone wiring laid out, eg do you have a master with multiple extensions coming off it in star format, or do you have extensions daisy-chained from the master? Is the router connected to an extension lead, do you have a Sky box.

If you can post your line stats, we'll have a better idea on what speeds you should be able to achieve.

[mattpearce79]

From what i can remember it was pretty steady at 3000 for the first month or so then dropped steadily down to 1856 ish since then.

My wiring....i have a filter at my main phone socket that i wired directly to my router via the adsl (10 metre) cable, I then connect to the router via a wireless adapter. Unfortunately i have a wierd phone socket that doesn't have a test socket, (i think Buzby fitted in back in the early 80's!!!)

My stats are:-
speed down 1440 kbps up 448 kbps
Line Attenuation down 63.5 db up 31.5 db
Noise Margin down 15.3 db up 16.0 db

[Rik]

TBH, Matt, I'm amazed that a 63.5db attenuation line ever got you a 3000 sync, it's doing pretty well as it is (on a fixed speed product, BT would only allow you 512k). However, you do appear to have a target margin of 15db. Were that 6db, you would have got the higher sync, but the margin has been increased in response to instability, which could explain the fall off you have seen.

If you're connecting wirelessly, you might benefit from using a shorter connection to the router, but without a test socket, there's not much else you can do for yourself. You say you have a filter there, is it a 'soap on a rope' type? How many extensions, if any, do you have and have you removed the ring wire from them?

It might be worth having a word with IDNet. They can test the line and, if necessary, get an engineer out to you. The risk is, though, that you could face a hefty bill if BT decide the problems is on your side of the master socket.

[mattpearce79]

My filter is a XF-1e Professional ADSL2/2+ Microfilter...i decided to pay a bit more to get a decent filter, as advised by Virgin that there could be a problem with my filter - it made no difference - turns out it was their network at fault! and no extensions just the ADSL and Phone cable plugged into the filter.

To be honest Rik i'm not concerned with mega download speeds i'm far more happier in the 3 days i've been with IDNET than i was for the whole 8 mnths i was with Virgin, pages load, connections are stable etc... but i just want to get the best speed possible on my line and just wanted to know if there was anything i could do about it or just leave it alone and hope it sorts itself out.

[Rik]

Hi Matt

What else is connected to the line, besides the router?

The xf-1e is, in my experience, the best filter on the market, but it's important to remember that the ADSL side is unfiltered, it's everything else that needs filtering.

Somewhere, you're almost certainly picking up noise, the issue is to find out where, and whether you can improve things.

[mattpearce79]

The only thing other than the router is the phone that does have an extension, there are a couple of old phone extension sockets in the house that are redundant and now wired to the master socket in any way.

The only other thing i can think of is that we have a wireless baby monitor in my daughters bedroom, which is where the router is also - could that be a factor?

[Rik]

It should only affect the wireless performance, Matt, depending on frequency band. Have you tried removing the ring wire?

[Si]

I was just about to suggest the same thing - I have an identical set-up to that described and removing my ring-wire made a noticeable improvement.

Having said that, if it doesn't help there's not an awful lot left to try with our "old school" BT sockets!

[mattpearce79]

Heres a pic of my wiring:-

[Rik]

You appear to have a split pair there, unless those two wire are the incoming line (in which case, where are the wire to the extension?).

[mattpearce79]

A little bit confused? those two wires come directly from my phone line outside the house, not too sure what you mean about the wire to the extension? Sorry am i being thick!

[Rik]

OK, if the wires are from outside, we don't have to worry about pairing. So is the phone on a 'loose' extension lead, not a hard-wired socket? If so, is that flat or round?

[mattpearce79]

i tried to type it...but found it easier to draw it!

[Rik]

So it's a loose extension, flat or round?

[mattpearce79]

Not quite sure what you mean Rik - sorry

[Rik]

You have an extension cable plugged into the filter, and the phone plugs into that, yes? Is that cable flat or round in its construction?

[mattpearce79]

Flat - sorry, got there in the end

[Rik]

That may be adding to your problems. Try and get a round one (B&Q usually have them). Flat cables do not use twisted pairs, round ones do. The twisting neutralises a lot of the noise pickup, so even though the cable is after the filter it's worth trying the cable (or just temporarily plug the phone into the filter directly) to see if the noise margin increases when the flat cable is removed.

[mattpearce79]

Could i not just unplug the phone and see if the noise margin increases or would it not be as immediete as that?

[Rik]

Unplugging the phone wouldn't eliminate the cable. However, if you unplugged the extension at the filter, you would establish whether there was any effect from it, and it would then be a case of narrowing the search if necessary.

[colirv]

i tried to type it...but found it easier to draw it!

Did you draw that yourself? If so, which program did you use?

<impressed>

[mattpearce79]

i used photoshop

[Rik]

Now, I'd have used Illustrator or Corel Draw...

[colirv]

Matt's a professional!

[Rik]

 

A professional would have used a drawing program, Colin.

[colirv]

A professional would have used a drawing program, Colin.

<pulls up chair, reaches for popcorn>

[Rik]