http://tbb.st/1575925770282583955 Suffolk - oddly high but think someone is testing something
http://tbb.st/1575228253291684655 Northampton
http://tbb.st/1575988787890146655 Birmingham
http://tbb.st/1576400355570292955 Cornwall
http://tbb.st/1577480280625667255 Edinburgh

Posters latest test http://tbb.st/1578302332575595355

The spiky nature means it is possible the ookla tester which tends to have a higher sampling rate is seeing some higher results and may be reporting those figures.

Dug through the recent Plusnet and http://tbb.st/1578304091143085155 shows what a good test with full sync should look like

Thanks Mr S, any TBB FTTP speedtests in the Highlands/North Scotland/Aberdeenshire? I've yet to see a 'perfect' one in these areas.

In reply to a post by RobertoS:

Basically on mobile broadband, Ookla shows peak, tbb shows what looks real from visual analysis of that other monitor.

Ookla doesn�t show the highest speed achieved within the speed test overall, however will manage the connection to try and maximise throughput. It takes the maximum speed achieved right at the end of the test to present back as the achieved figure. This could be lower than the maximum seen within the test itself.

For anyone interested this is how Ookla speed test operates for download:

The client establishes multiple connections with the server over port: 8080. The client requests the server to send an initial chunk of data.
The client calculates the real-time speed of the transfers, then adjusts the chunk size and buffer size based on this calculation to maximize usage of the network connection.
As the chunks are received by the client, the client will request more chunks throughout the duration of the test.
During the first half of the test, the client will establish extra connections to the server if it determines additional threads are required to more accurately measure the download speed.
The test ends once the configured amount of time has been reached.

In reply to a post by gary333:

Ookla doesn�t show the highest speed achieved within the speed test overall, however will manage the connection to try and maximise throughput.

Easily tested too. Just start an Ookla speedtest and begin another download half way through, the end result will be significantly below peak.

The myth that the Ookla speedtest result indicates peak speed has been propagated for a long time.

No need to be rude.

I was replying to your post, where you were implying that the TBB tester cannot give false results.

Also I don't see why I've "hijacked" the thread when the last paragraph is totally relevant to the subject title.

In reply to a post by baby_frogmella:

The point I'm trying to make is that results from the TBB tester shouldn't always be treated as gospel, especially since you have loads of different Ookla servers to compare against and perhaps more importantly, real life file download speeds as well. No way on earth would I be getting a constant 40 MB/s file downloads if the line speed was only 150 Mb/s.

Edited by deleted (Mon 06-Jan-20 15:40:54)

Have tried that and I am seeing different behaviour that supports the peak speed theory

Have also seen it on other peoples connections when watching video recorded of the test

The theory that state is that 25% of the slow samples and 15% of the fast samples are rejected and then an average calculated, but that does not seem to be the case all the time.

If I dig through all the ISP probably

Vodafone Gigafast Stirling
http://tbb.st/1575362642381874855

We don't have any different peering for regions, so a BT customer in Edinburgh should not be significantly different from someone further north.

I've looked and don't see where he says the tbb tester cannot give false results

All speed testers even hardware based ones are prone to vagaries of the Internet and other issues that may produce odd results. The

Your post, whilst doubtless factually valid wrt to your own experience, is not valid as a useful response to the OP. Read the opening post again please, think carefully about it, and decide which of the two tests seems more to reflect the reason for the poster conducting the tests in the first place.

I am not saying either is accurate. But your post did hijack the thread in favour of what appears over a period to be a hobby-horse of yours.

How about you think about what might be causing the OP's real-use problem with throughput? Ignore the tests if you wish.