Which is the most accurate Braodband speedtest site

  Kevscar1 03:14 29 Sep 12

because they seem to differ wildly so I ran an experiment starting four going as soon after each other as I could the dong it in the reverse order. None of the results match another.

http://www.mybroadbandspeed.co.uk/ 44.14 43.43

http://www.speedtest.net/ 31.41 27.94

http://www.uswitch.com/broadband/speedtest/?utmmedium=cpc&utmsource=bing&utmcampaign=Bing+BB+%28Search+Network%29&utmcontent=Broadband+speedtest+%28all%29#speedtestID=22923394&products=postcode%3DMK6%2B4HH%26advanced%3D1 15 48.7

http://www.broadbandspeedchecker.co.uk/ 8.34 14 .92

  wiz-king 08:10 29 Sep 12

Difficult to tell what your speed is - It can vary by the minute. Different speed test sites use geographically different servers and paths, I prefer speedtest.net as I can choose the server (I prefer to use one several hundred miles away) most of the others you have no idea of the path length. You can only do comparative tests and if you are going to complain to your ISP you will need to use the one they recommend.

Speed testing results can be skewed by using a wireless connection or a home-plug network, even a change of router.

  Kevscar1 08:24 29 Sep 12

I prefer to use one several hundred miles away)

but if thats not the one you are normally connected through then you are getting a false figure.

  Picklefactory 09:02 29 Sep 12

It varies wildly, i asked exactly the same question in the Helproom a month or so ago. There wasn't a conclusive answer other than people stating their preferred site.

  Picklefactory 09:08 29 Sep 12

I think the best advice i found was on the Virgin site, where it recommends selecting 3 or 4 hefty downloads from a reliable source (Reliable in terms of speed consistency) downloading them simultaneously, and simply timing the downloads. That would be, i think, as accurate as possible, and could then be used as a benchmark for speed test sites.

  wiz-king 09:33 29 Sep 12

Your internet always goes to your ISPs nearest node - mine is about 10 miles away - and then is sent on to its destination and may go through many other nodes all can introduce time delays so every website you call will have a different speed.

There is no such thing as a definitive speed, but by choosing a constant destination you are tending to remove one variable. By using several varied sites you just confuse the issue as they will be using different servers - one may be in London and another in Edinburgh.

  Picklefactory 10:02 29 Sep 12

wiz-king I understand what your saying, but all that achieves is consistency of results, not accuracy. I'll dig out a link to my thread in the Helproom, which showed a huge difference in speed between sites. I could quite easily pick one of those and get a consistent result, but it differed hugely from the equally consistent result from a different site. I suppose maybe i was asking the wrong question. You rightly point out that speed may vary dependent on node location, so maybe the question should be, how do i measure the maximum speed i can receive?

  Picklefactory 10:13 29 Sep 12

click here

This is the thread from the Helproom

  Forum Editor 10:41 29 Sep 12

This should be in the Helproom. I'm transferring it from Speakers Corner now.

  wiz-king 10:51 29 Sep 12

Try Dan Elwells program - the free bit - it is more accurate than a web-based checker I have used it for years.

  wiz-king 11:00 29 Sep 12

Drat - link not working properly - right program but it cant find the server list.

This thread is now locked and can not be replied to.

Huawei P10 review

1995-2015: How technology has changed the world in 20 years

An overview: What leading creative agencies are doing to improve diversity

New iPad, iPhone SE & Red iPhone 7 on sale now