Interenet speed, I know , lets make it stupidly complicated.

[Me!]

Sorry about hijacking your thread Rob, but I think we are fighting a common enemy here, ... BT!

I didn't know about pathping, but trying it gives me this -


That seems to be telling me that there are several places where I'm getting 100% loss! But the first of them looks like it is one hop further on that the IP address that I suspected, is that correct?

 

[MdB]

Ignore the 100% loss. These are computers along the route which don't respond to pings.
The interesting bit is the 100+ms ping times which are very poor. Hop 2 is the first machine outside of your home network and this has the poor ping time so this is more evidence that the problem is between your exchange and your ISP.

As a comparison, my ping time to telehouse (500 mile round trip to Cornwall) is 29ms.

 

[Me!]

No LLU available, no ADSL2+
BT have just phoned me though, they "have just tested my line and it all looks fine" :evil:
Has anyone tried domestic satellite broadband in the UK? Or, in France they have very good broadband via some kind of RF and an ordinary antenna on the roof, but I can't find anything like that in the UK for domestic use, only for commercial. Does anybody know of a supplier please?

 

[Penners]

MdB - I knew about ping, but not pathping. Very interesting! I've just tried it to bbc.co.uk and got 33ms. What puzzles me is the fact that some of the later stage time readings are lower than earlier ones. Me!'s readings show the same.

If I understand it correctly, the time recorded at each of the stages (10 in my case) is from the source (ie: my computer) to that point, and not from the previous point to that point, as the column is headed "Source to here". Is that correct? If so, wouldn't you logically expect the time measured at each stage to be incremental?

And yet this isn't the case.

 

[MdB]

BT have just phoned me though, they "have just tested my line and it all looks fine" :evil:

No surprise there as there isn't anything wrong with your line - your good sync speed shows that. It is the exchange that has the issues. Line tests won't detect this at all.

I would suggest posting your issues here: http://forums.thinkbroadband.com/ There are some good people there with more knowledge than me. They may well be able to tell you the correct wording needed to get your problem looked into properly.

 

[MdB]

If I understand it correctly, the time recorded at each of the stages (10 in my case) is from the source (ie: my computer) to that point, and not from the previous point to that point, as the column is headed "Source to here". Is that correct? If so, wouldn't you logically expect the time measured at each stage to be incremental?

And yet this isn't the case.

Weird isn't it? The bit of info you are missing is that different bits of kit prioritise ping requests differently. Some of the nearer kit has put it at a low priority so take longer to respond than kit further down the chain.

 

[Feltwell]

BBC News reveals a startling fact today - "A third of homes in the UK have broadband speeds well below the national average"

Who would have thought it - that if you look at all the figures that together make an average, some of them are below that average, some are even well below that average.
There's a startling revelation. :roll:

 

[Flyfisher]

Or, more positively, two-thirds of homes have faster than average broadband.

 

[Feltwell]

It was a third that were "well" below. So you never know, it may be that roughly half are below average - or possibly even exactly half if they used the median as their average. That would be a shock.

 

[LadyArowana]

To drag this vaguely back on topic. I saw my friend with the broadband problems yesterday and she reported that it was FINALLY fixed and giving proper speeds. After months of customer support uselessness an many versions of what was causing it, an engineer was sent. As far as we could work out it was a combination of an old socket and interference from a Sky box. He changed some "plastic bits on the wall" and it's now fine. I have no idea what the technical reasons for this were, but I pass it along in case it is of any use to anyone else!

 

[Moo]

I'm sorry to sound like Mrs Smug, but our broadband speed is excellent as is the service we get from our provider. Last week one of our digi-boxes died of old age and its replacement was installed the same day.

 

[Penners]

If I understand it correctly, the time recorded at each of the stages (10 in my case) is from the source (ie: my computer) to that point, and not from the previous point to that point, as the column is headed "Source to here". Is that correct? If so, wouldn't you logically expect the time measured at each stage to be incremental?

And yet this isn't the case.

Weird isn't it? The bit of info you are missing is that different bits of kit prioritise ping requests differently. Some of the nearer kit has put it at a low priority so take longer to respond than kit further down the chain.

Thanks for that explanation, MdB. I'm deeply wounded that anyone should regard my business as low priority.

 

[Penners]

Or, more positively, two-thirds of homes have faster than average broadband.

Hang on! If 2/3rds are above average and 1/3rd is below average, doesn't that mean that the average figure must be wrong?

My head hurts...

 

[Penners]

I'm sorry to sound like Mrs Smug, but our broadband speed is excellent as is the service we get from our provider.

Thass coz you're in the big ci'y, missis!

 

[Feltwell]

Or, more positively, two-thirds of homes have faster than average broadband.

Hang on! If 2/3rds are above average and 1/3rd is below average, doesn't that mean that the average figure must be wrong?

My head hurts...

No - at least not if you regard the average as being the "mean", which most people do, rather than the median (middle figure of all the data if it's put in order) or the mode (most frequently occuring figure) which can also legitimately be called the average.

2/3 of households could be slightly above, and 1/3 a long way below, the mean average.

So the average can be one of 3 totally different numbers and still be correct. What was it they said about lies, damned lies and statistics?

All of which isn't what the Beeb said anyway!

*I hope the above's right. It's been a long, long time since I studied maths!

 

[Penners]

My head still hurts.

 

[Feltwell]

That'll be last nights whisky.

 

[Penners]

Incorrect.

Now, if you'd said gin...

 

[Flyfisher]

*I hope the above's right. It's been a long, long time since I studied maths!

Me too, because I was going to say the same thing, but you beat me to it.

Example: 1/3rd get 2Mbps, 2/3rds get 8Mbps Average speed is 2+8+8 = 18 / 3 = 6 Mbps.

1/3rd get "well" below the average of 6Mbps leaving the other 2/3rds being above average.


Statistics never actually 'lie', though they can throw up some counter-intuitive results. For example, I would bet a decent sum of money that I have more than the UK national average number of legs:

UK population: 65m (say)
Number of legs: 65m x2 minus a small number of missing legs due to unfortunate accidents, medical conditions etc. Let's say 1000 for illustration.

So, UK national average number of legs per person = 130,000,000 - 1000 = 129,999,000 legs / 65m people = 1.999984615

I happen to have two legs, thus I have slightly more than the UK national average - and so do the vast majority of the UK population.

 

[Penners]

I happen to have two legs, thus I have slightly more than the UK national average - and so do the vast majority of the UK population.

:lol:

It's a statistical fact that 97% of red-haired people eat potatoes. Therefore potatoes definitely cause red hair.