ChatterBank3 mins ago
Answers
Best Answer
No best answer has yet been selected by joko. Once a best answer has been selected, it will be shown here.
For more on marking an answer as the "Best Answer", please visit our FAQ.
Here's another site:
http://www.speedtest.net/
I like that one because it interprets the results by telling you how long it will take to upload or download different types of file.
It's also a good idea to check the quality of your phone line by running a ping test:
http://www.pingtest.net/
(You're looking for an A or B grade).
Chris
http://www.speedtest.net/
I like that one because it interprets the results by telling you how long it will take to upload or download different types of file.
It's also a good idea to check the quality of your phone line by running a ping test:
http://www.pingtest.net/
(You're looking for an A or B grade).
Chris
Ignore all of these speed "test" sites - they are all, without exception total garbage! Think about it - how can they possibly work. All that they can possibly tell you is how quick your connection is between YOUR computer and THEIR server. Hundreds of variables can affect that. Do a trace to speedtest.net and then laugh a lot...
The ONLY way to measure your true connectivity is to download sufficient numbers of files from your ISP local network until your bandwidth is maxed out, and then add the individual speeds up...
The ONLY way to measure your true connectivity is to download sufficient numbers of files from your ISP local network until your bandwidth is maxed out, and then add the individual speeds up...
Errr, Mark,
You can have the fastest connection in the world to your ISPs network, but if their upstream peering policy is crap then it's no good at all, you want speed to the world, not just to your ISP.
A test between your computer and a computer outside of your ISPs network is a much more relevant real world test.
You can have the fastest connection in the world to your ISPs network, but if their upstream peering policy is crap then it's no good at all, you want speed to the world, not just to your ISP.
A test between your computer and a computer outside of your ISPs network is a much more relevant real world test.
That's one of the reasons, Mark, that speedtest.net uses multiple servers (and allows you to switch between them). It's also a good reason for using several different test sites. Yes, they have to make certain assumptions about the efficiency of the 'backbone' path between the test server and your local telephone exchange but it's usually the very end bit of the path (i.e. between the exchange and your computer) where any real bottlenecks occur.
Just down the road from me (in Kesgrave, Suffolk), BT are about to trial their 1000Mbps internet service. There wouldn't be much point in offering such a service into people's homes if the 'backbone' couldn't support such speeds for the route between the host server and BT's equipment. Sites like speedtest.net simply make the same assumption that BT does (i.e. that there's a 'clear path' on the backbone).
Chris
Just down the road from me (in Kesgrave, Suffolk), BT are about to trial their 1000Mbps internet service. There wouldn't be much point in offering such a service into people's homes if the 'backbone' couldn't support such speeds for the route between the host server and BT's equipment. Sites like speedtest.net simply make the same assumption that BT does (i.e. that there's a 'clear path' on the backbone).
Chris
Then you get a slow connection....just as you would in the real world if the same happened.
I will concede that a single test is going to give no more than a snapshot and there is no way you can base any assumption of speed on one test, but running tests solely between your computer and your ISPs network is testing your ISPs network which in my book is not an internet connection.
On a few occasions the ISP we use where I work has lost all connectivity outside telehouse, during these periods we are still able to access any site hosted inside telehouse on our ISPs network at lighting speed but we have zero access to any sites outside telehouse, by your method this means we still have a good, fast internet connection.
I will concede that a single test is going to give no more than a snapshot and there is no way you can base any assumption of speed on one test, but running tests solely between your computer and your ISPs network is testing your ISPs network which in my book is not an internet connection.
On a few occasions the ISP we use where I work has lost all connectivity outside telehouse, during these periods we are still able to access any site hosted inside telehouse on our ISPs network at lighting speed but we have zero access to any sites outside telehouse, by your method this means we still have a good, fast internet connection.
-- answer removed --
Try this link, Chuck. (It's routed via a proxy server):
http://tinyurl.com/2wxt7ps
http://tinyurl.com/2wxt7ps
> What we found was surprising – we always suspected that some speed tests out there > weren’t able to keep up with our speeds, but we never realised just how bad some of
> them are.
But how can this possibly be? CF is an internationally recognised IT expert. VirginMedia must surely have got it wrong... ;-)
> them are.
But how can this possibly be? CF is an internationally recognised IT expert. VirginMedia must surely have got it wrong... ;-)
Cheers Chris..
OK that link seems to be about bandwidth throttling, which I will grant you I didn't consider (as it doesn't apply to me). But I maintain that performing tests between you and your ISPs network is testing your ISPs network more than your internet connection.
Even if your ISP is throttling you to a lower rate then that is going to be an indication of the real world speed you are likely to get when doing a http download which is what most people will consider is their internet connection.
If you were to run a test between you and your ISP and could get 20Mb/s all day everyday but as soon as you try to download anything from any other site outside your ISPs network you only got 512Kb/s would you consider you had a real world usable 20Mb/s connection?
OK that link seems to be about bandwidth throttling, which I will grant you I didn't consider (as it doesn't apply to me). But I maintain that performing tests between you and your ISPs network is testing your ISPs network more than your internet connection.
Even if your ISP is throttling you to a lower rate then that is going to be an indication of the real world speed you are likely to get when doing a http download which is what most people will consider is their internet connection.
If you were to run a test between you and your ISP and could get 20Mb/s all day everyday but as soon as you try to download anything from any other site outside your ISPs network you only got 512Kb/s would you consider you had a real world usable 20Mb/s connection?
Latest Posts
Crosswords7 mins ago
Shopping & Style8 mins ago
Crosswords9 mins ago
Crosswords10 mins ago
ChatterBank12 mins ago
