Broadband speed tests
Thread Starter
Broadband speed tests
Living, as we do, in a rural location some 4miles from civilisation, we get a very poor broadband speed. It seems that BTare unable to persuade high speed electrical bits to travel down 4 miles ofcopper cable when all their mates have the luxury of fibre optic travel. However,I digress, sampling 3 random broadband speed test sites they all (incredibly) comeback with the same answer of 3.4 megathings. Two sites completed the test in 12seconds, the third took 30 seconds. In the middle of the test the ‘clock’ of the slow one stuck at 50% for ages. The cynic in me smells something ratty here. Are they saying that during actual download the speed is indeed 3.4 but there is an awful lot ofwaiting time between the packets – but we don’t count that sir because it’s beyond our control?
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Distance from your cabinet is crucial too in the information you provide. How far (by road) are you? Speed tests are notoriously unreliable. What does the BT Wholesale checker say about your line?
Join Date: Feb 2008
Location: High seas
Posts: 216
Likes: 0
Received 0 Likes
on
0 Posts
Broadband speed tests
Had satellite internet fitted for the same reason - we only getting 0.8 mb/s due to cable run distance. Total outlay £800 fitting plus £49 p/m.
However, now the speed test says we are getting 19.6 mb/s, but it takes 3 times as long to download stuff!
Go figure that one!!
I am not too impressed.
Speed test. Bah.
However, now the speed test says we are getting 19.6 mb/s, but it takes 3 times as long to download stuff!
Go figure that one!!
I am not too impressed.
Speed test. Bah.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Squeegee - satellite is known to drop to 1 or 2mb at peak times (do not believe the 'hype') and you could be getting a load or errors on the connection too.
Join Date: Aug 2002
Location: Earth
Posts: 3,663
Likes: 0
Received 0 Likes
on
0 Posts
Speed tests are notoriously unreliable.
As I say time and time again on here, its not about the speed, its about the quality of service. Two broadband services which are identical on paper can have a vastly different feel.
Had satellite internet fitted for the same reason - we only getting 0.8 mb/s due to cable run distance. Total outlay £800 fitting plus £49 p/m.
However, now the speed test says we are getting 19.6 mb/s, but it takes 3 times as long to download stuff!
However, now the speed test says we are getting 19.6 mb/s, but it takes 3 times as long to download stuff!
Join Date: Jan 2008
Location: Bracknell, Berks, UK
Age: 52
Posts: 1,133
Likes: 0
Received 0 Likes
on
0 Posts
Satellite broadband is almost always not as fast as you'd think it is because you're not including the latency aspect into things. i.e. it's comparing oranges and apples.
To calculate the expected throughput of a link, see this link
How to Calculate TCP throughput for long distance WAN links
To calculate the expected throughput of a link, see this link
How to Calculate TCP throughput for long distance WAN links
Join Date: Dec 2013
Age: 81
Posts: 316
Likes: 0
Received 0 Likes
on
0 Posts
A speed of 3.4 megathings is indeed slow and will slow up your browsing something rotten. People in towns and reasonably near the exchange usually get up to 20 megathings.
But the total website browsing experience is the sum of all the system delays, and these include how good the website is, how many users are loading it with requests, how well the message packets make it to your provider, and how well the provider's infrastructure dishes out everything to you and your fellow subscribers.
It is not how fast your line is that matters, it is how all the slownesses add up.
But the total website browsing experience is the sum of all the system delays, and these include how good the website is, how many users are loading it with requests, how well the message packets make it to your provider, and how well the provider's infrastructure dishes out everything to you and your fellow subscribers.
It is not how fast your line is that matters, it is how all the slownesses add up.
One's internet has an average speed of 2.2 megasomethings, really rather pedestrian in these modern times, although at least the last 50 metres is through overhead wires and a good mile from the exchange. And yet one's internettery is for the most part acceptable. Videos at the higher resolutions are better left alone of course. It truly is a baffling experience trying to understand the black arts of internet speed what with contention, congestion etc etc.
SHJ
SHJ
Avoid imitations
Join Date: Nov 2000
Location: Wandering the FIR and cyberspace often at highly unsociable times
Posts: 14,576
Received 429 Likes
on
226 Posts
I suffered for years with a slow or non-existent broadband connection and my complaints to the provider then to BT only resulted in a statement that my home location was too far from the exchange for it to ever work properly. The connection was often too slow for the Speedtest page to load - it kept timing out.
I now have the luxury of a 38Mb/9Mb Infinity connection via the very same line.
I recently carried out a series of speedtests and tried a internet provider service comparison. It tells me that I have the fastest connection in the area! No idea why or how but I'm not complaining!
Having said this, over last weekend the connection kept dropping out completely. It's back up again today.
I now have the luxury of a 38Mb/9Mb Infinity connection via the very same line.
I recently carried out a series of speedtests and tried a internet provider service comparison. It tells me that I have the fastest connection in the area! No idea why or how but I'm not complaining!
Having said this, over last weekend the connection kept dropping out completely. It's back up again today.
Advertised sownload rate or speed is much like fuel economy. yes you can get 58mpg with a 9 st driver maintaining exactly 56mph on a dry road with no wind and nothing in the car at all except about 2 gallons of gas and maybe even with a few non essential bits taken off.
I live very close to my local BT exchange-cable route-fibre optic to the curb is 300m and I get 76mbs . BT advertise 80 but you would only get that in the exchange itself .
When trying to find out what really works you need to find a fairly stable site that you can download something from - say a video clip or tune where it tells you the file size. If you know where the server is you can run a test download at different times of the day which will take out the contention issues if its an unsocial hour with you and the server. 0600 uk is 0100 EST in USA so you are likely to get a better result than at say 1900 uk 1400 USA.
And never try any test when its close to midnight in California or US Central time since these are the times when all sorts of little fixes and mods and changes get done on the network so consistency isn't going to happen.
If you run your download time at the off peak times suggested then you can in parallel run a BT speed test or Ooklas one and see what speeds they show at that time of day and you can compare that with the actual download to give you some idea of the gap between advertised mpg and what happens in your car day to day.
If you do get consistently low speeds go back to your ISP and complain and explain you have tested it in the wee hours of the night and compared it with downloads done at quiet hours. There are all kinds of things that can go wrong with the local loop ( exchange to your house) bit especially if you do not use BT. BT still run all the cables so everything comes down to them and they only give the actual ISP outline info on loop distance and performance and this can easily be wrong due to changes in cabling or equipment in the BT part of it which do not get transferred to the ISP.
BY the way overhead cable or underground shouldn't really make any difference. But satellite is a basically a dead loss , data protocols do not like the delay inherent in satellite connections and seldom work well which is why satellite are not used for international communications anymore.
Your ISP can change the line speeds and some of the line characteristics from the defaults BT give them and that can be worth a try and the route a cable takes to your house can be miles-literally-different to what you see above ground. Some years ago I swapped to O2 and got slow speeds which they said was because the cable route to my house was 3000m whereas it is actually 300m, either e typo in the info BT gave to O2 or the fact that BT altered the cables at some point.
So its real problem if you don't live close (2500m or less ) unless you use BT Infinity which terminates at the nearest cabinet (green box on pavement) to you.
Sorry to ramble on a bit but I have found that a bit of research pays dividends on this issue and not taking on trust what either BT or your ISP say without challenge
Good luck
I live very close to my local BT exchange-cable route-fibre optic to the curb is 300m and I get 76mbs . BT advertise 80 but you would only get that in the exchange itself .
When trying to find out what really works you need to find a fairly stable site that you can download something from - say a video clip or tune where it tells you the file size. If you know where the server is you can run a test download at different times of the day which will take out the contention issues if its an unsocial hour with you and the server. 0600 uk is 0100 EST in USA so you are likely to get a better result than at say 1900 uk 1400 USA.
And never try any test when its close to midnight in California or US Central time since these are the times when all sorts of little fixes and mods and changes get done on the network so consistency isn't going to happen.
If you run your download time at the off peak times suggested then you can in parallel run a BT speed test or Ooklas one and see what speeds they show at that time of day and you can compare that with the actual download to give you some idea of the gap between advertised mpg and what happens in your car day to day.
If you do get consistently low speeds go back to your ISP and complain and explain you have tested it in the wee hours of the night and compared it with downloads done at quiet hours. There are all kinds of things that can go wrong with the local loop ( exchange to your house) bit especially if you do not use BT. BT still run all the cables so everything comes down to them and they only give the actual ISP outline info on loop distance and performance and this can easily be wrong due to changes in cabling or equipment in the BT part of it which do not get transferred to the ISP.
BY the way overhead cable or underground shouldn't really make any difference. But satellite is a basically a dead loss , data protocols do not like the delay inherent in satellite connections and seldom work well which is why satellite are not used for international communications anymore.
Your ISP can change the line speeds and some of the line characteristics from the defaults BT give them and that can be worth a try and the route a cable takes to your house can be miles-literally-different to what you see above ground. Some years ago I swapped to O2 and got slow speeds which they said was because the cable route to my house was 3000m whereas it is actually 300m, either e typo in the info BT gave to O2 or the fact that BT altered the cables at some point.
So its real problem if you don't live close (2500m or less ) unless you use BT Infinity which terminates at the nearest cabinet (green box on pavement) to you.
Sorry to ramble on a bit but I have found that a bit of research pays dividends on this issue and not taking on trust what either BT or your ISP say without challenge
Good luck
Controversial, moi?
Sorry to ramble on a bit but I have found that a bit of research pays dividends on this issue and not taking on trust what either BT or your ISP say without challenge
What a novel concept, I doubt it will catch on.
Join Date: Jan 2008
Location: Bracknell, Berks, UK
Age: 52
Posts: 1,133
Likes: 0
Received 0 Likes
on
0 Posts
No you wouldn't. 76mbit/s is the profile maximum, so whilst they say 80mbit/s it's actually a bit of misadvertising...but only by 4mbit/s. Equally, you'll only get circa 18mbit/s upstream too.
Join Date: Dec 2013
Age: 81
Posts: 316
Likes: 0
Received 0 Likes
on
0 Posts
One thing that seems to contribute to slowness is the modern fashion for as-you-go suggestions. Way back in time you typed something into googoo and hit enter to get a reply. Now it keeps suggesting things every time you hit a key. This must be very wasteful. Even worse is the discus (spelt with a q) comments service used by the Telegraph. Wish you could switch off the smarts and maximise efficiency by keeping the system dumb.
Join Date: Jan 2008
Location: Bracknell, Berks, UK
Age: 52
Posts: 1,133
Likes: 0
Received 0 Likes
on
0 Posts
One thing that seems to contribute to slowness is the modern fashion for as-you-go suggestions. Way back in time you typed something into googoo and hit enter to get a reply. Now it keeps suggesting things every time you hit a key. This must be very wasteful. Even worse is the discus (spelt with a q) comments service used by the Telegraph. Wish you could switch off the smarts and maximise efficiency by keeping the system dumb.
One of the biggest culprits of slowing things down, albeit your PC and not your broadband service, is the Skype Click-to-call service. This is a service which ostensibly underlines clickable phone numbers in web pages. However, whilst that service has dubious benefits, it has one huge drawback as the service is so badly written it slows down most computers that have Skype installed. I would suggest anyone using Skype uninstalls this service.
I'm always a bit suspicious with speedtests, but they do have their place. I recently returned a repaired PC to a remote house. Download speed seemed dire, so ran speedtest.net. 0.04Mb ! Pluigged the router into the test point on the master socket and got 3.5. House extension wiring can be troublesome.
Mike Bracknell
Thanks for the interesting point there-I was aware that the speed can be limited by the provider but I did not know that BT actually limit the speed to below 80gbs so that it is literally unattainable under any circumstances. Obviously a few exemptions from advertising laws in play there. Like my car analogy its bit like saying the engine does 80mpg but not with a car attached.
PB
Thanks for the interesting point there-I was aware that the speed can be limited by the provider but I did not know that BT actually limit the speed to below 80gbs so that it is literally unattainable under any circumstances. Obviously a few exemptions from advertising laws in play there. Like my car analogy its bit like saying the engine does 80mpg but not with a car attached.
PB
Join Date: Aug 2002
Location: Earth
Posts: 3,663
Likes: 0
Received 0 Likes
on
0 Posts
Obviously a few exemptions from advertising laws in play there.
