Speed Test results

Hi,
Are there big differences in the thinkbroadband speedtest and speedtest.net ?

Here shows around 552.85 Mbps
https://www.thinkbroadband.com/speedtest/17384208642...

speedtest.net around 940 Mbps.
This test was slightly slower:
https://www.speedtest.net/result/17324889343

Downloading a Linux ISO via torrent and it was recording the speed at 96.5Mib/s

Is there way to get an accurate speed test ?

Thanks

In reply to a post by zMaliz:

Is there way to get an accurate speed test ?

They are all accurate in that they all show the speed between their location and yours. And dependent upon ISP and users' geographic locations, different speed testers may turn out to be more accurate for some more than others.

What you want is a speed tester which can saturate your line - if it is the capability of your line you want to know.

You could try fast.com

In reply to a post by zMaliz:

Is there way to get an accurate speed test ?

The speed tests are as accurate as the route to the host, so the speed you get from Thinkbroadband is the max speed your ISP was able to get you to that host. Others can get faster to Thinkbroadband. (The owners have demonstrated their speed test can go very fast)

If using Windows or macOS you can use TaskManager or Activity Monitor when downloading a large file, for gigabit you will need a very very large file; and check the data rate. Win10 and Win11's task manager make this easy.

If you are on Windows removing (not disabling) non Microsoft security products may improve speed; many take over the network stack.

As said, you can’t take any one speed tester, app or indeed a particular server that they point to as some gospel or “the best”…

As someone remarked on here recently, the whole internet is best efforts. Very true.

Take a “basket” approach. Look at the mean average and judge from that viewpoint rather than a single ‘great’ test (or single terrible test).

Edited by Pheasant (Sat 01-Feb-25 21:13:49)

In addition to what others have noted it would be worth knowing what you are trying to find out. If you are trying to find out how long it will take to download something then the only valid test is to download it. There is after all no point knowing how long it will take you to drive to Edinburgh if you, in fact, want to go to Liverpool.

If you want to know the fastest speed your line can handle then that could be tricky. The only valid test there would probably be to use a tester based at your telephone exchange. Or maybe even one based in your cabinet (for FTTC) or aggregation node (FTTP). But in all those cases the result wouldn't be all that helpful either because that's like knowing how quickly you can get out of your housing estate.

The best use of speed testers is to run them fairly frequently and learn to know what they typically report. That way you can spot faults if they don't report what they normally do. The Thinkbroadband tester is particularly good because it reports both single threaded and multi-threaded with graphs.

Of course even then the tester on its own doesn't tell you where the fault is but at least you can post the result somewhere on here (preferably your ISP's section) and if other people say they see the same thing you can perhaps rule out your equipment being the cause and use the responses as evidence when talking to support.

Edited by Andrue (Sat 01-Feb-25 21:14:26)

Thanks for the replies.

I just wanted to be sure I was capable of getting the speeds I'm paying for.
I was also curious about the differences seen with different speed tests.

All the comments make sense.
Thanks

In reply to a post by Pheasant:

Take a “basket” approach. Look at the mean average and judge from that viewpoint rather than a single ‘great’ test (or single terrible test).

The median, max and third quartile are good to know. I find the average is not much use. If I get a Mean of 475.53 mb/s (actual figure from my router's monitoring for the past hour period) then what does that mean exactly? Is my Internet working fine or not? I can't tell.

If I have a median of 659.86 mb/s then that's much more useful, I know immediately that half of my speed tests lie to the left of this and the other half to right of this. This is a great ball park figure.

If I have a 3rd quartile of 793.63 mb/s and a max of 854.02 mb/s then it becomes immediately obvious that my Internet is fine and I don't have to go complaining for compensation.

Also worth knowing that some speedtests have been known to show speeds well in excess of what a line is capable of. Different speedtests use different algorithms and some are affected by the specific browser and OS they are running on.

Yes very mathematically precise. Possibly very slightly OTT too 😉

What I was saying to the OP was simply run a basket of tests at different times from different places. You can pretty quickly get a feel for what is about right for your connection, but by all means you can log all this in a spreadsheet and figure out the third quartile and the other stuff if you want to get quite scientific about the whole thing. I don’t think most folks will necessarily go that far…

Edited by Pheasant (Mon 03-Feb-25 21:37:39)

In reply to a post by zMaliz:

Hi,
Are there big differences in the thinkbroadband speedtest and speedtest.net ?

Here shows around 552.85 Mbps
https://www.thinkbroadband.com/speedtest/17384208642...

speedtest.net around 940 Mbps.
This test was slightly slower:
https://www.speedtest.net/result/17324889343

Downloading a Linux ISO via torrent and it was recording the speed at 96.5Mib/s

Is there way to get an accurate speed test ?

Thanks

I use samknows speedtest very accurate! https://embeddable-speedtest.samknows.com/

For info - I'm on Swish 1000 and I don't think there are any contention issues in my area thankfully.

On Fast.com I nearly always get 1gbit down and >950mbit up
Most others are around the 920-950mbit range.

e.g.
Speedtest 20250207

I'm on a M1 Mac using Brave browser but it seems similar on windows 10/11 when I have tried.

In reply to a post by smouty:

On Fast.com I nearly always get 1gbit down and >950mbit up

If you have a top end corporate network and have a fast server and your PC connected to an enterprise switch with gigabit connection, the fastest you see is 943 Mbit. So your web based speed testers giving higher than this are reporting something else, perhaps caching in your browser, or security software or the network.

or just trying to give you a high number for bragging rights.

It has already been mentioned that fast.com reporting is a bit off.

I'm not saying is correct but just what it is e.g. another comparable reference point

In reply to a post by smouty:

It has already been mentioned that fast.com reporting is a bit off.

has been for years, it was created by Netflix so you could see if you had enough capacity to watch Netflix content.

Fiction just like its creator 😂

In reply to a post by Pheasant:

Fiction just like its creator 😂

It does seem "to depend". I've found it works okay on a Windows PC using Firefox in a "private window", but as Seb and Andrew (MrSaffron) have debated in TTTS the Firefox on Mac can behave quite differently in caching.

I believe the Thinkbroadband tester is designed to allow for these browser quirks, but the likes of fast.com are expecting to run on a mobile device over 3G (UMTS) 4G (LTE) or 5G (NR) connections, rather than more consistent WiFi.

In reply to a post by jchamier:

In reply to a post by Pheasant:

Fiction just like its creator 😂

It does seem "to depend". I've found it works okay on a Windows PC using Firefox in a "private window", but as Seb and Andrew (MrSaffron) have debated in TTTS the Firefox on Mac can behave quite differently in caching.

I believe the Thinkbroadband tester is designed to allow for these browser quirks, but the likes of fast.com are expecting to run on a mobile device over 3G (UMTS) 4G (LTE) or 5G (NR) connections, rather than more consistent WiFi.

I don't think any browser-based test are that reliable. Certainly anything like fast.com which (at least used to) over-read what what physically possible. Like telling you can get 1.2 Gbps throughput on a GbE port that can only physically sustain ~943 Gbps with the usual overheads.

Using the Speedtest.net app or command line tools seem to remove browser based artefacts. Not perfect though by any means, but better than a browser based test.

Saying that I can still kill the TTB speed tester - doesn't seem to cope very well at all above about ~3 GBps, and caps it at that...

In reply to a post by Pheasant:

Not perfect though by any means, but better than a browser based test.

I'm a fan of wget or curl a super large file and timing it, but that usually is only single connection so in modern WAN networks it appears you need multiple connections to show total throughput.

Saying that I can still kill the TTB speed tester - doesn't seem to cope very well at all above about ~3 GBps, and caps it at that...

I would expect that is deliberate

Probs. It's a bit tired to be honest. With the rise of XGS based AltNets there's enough folks out there with 2, 3, 5, 8 Gbps based broadband services.

Like the rest of this site, it needs to be brought well up to date, as its feels pretty creaky...

In reply to a post by Pheasant:

Probs. It's a bit tired to be honest. With the rise of XGS based AltNets there's enough folks out there with 2, 3, 5, 8 Gbps based broadband services.

ISPreview reported the actual takeup of services above 2 Gbps was miniscule in domestic. I know YouFibre is the 8Gbps leader but their coverage is tiny.

Like the rest of this site, it needs to be brought well up to date, as its feels pretty creaky...

Sadly yes, I assume the "day job" where the owners earn money has taken more of their time than before.

In reply to a post by jchamier:

In reply to a post by Pheasant:

Probs. It's a bit tired to be honest. With the rise of XGS based AltNets there's enough folks out there with 2, 3, 5, 8 Gbps based broadband services.

ISPreview reported the actual takeup of services above 2 Gbps was miniscule in domestic. I know YouFibre is the 8Gbps leader but their coverage is tiny.

Like the rest of this site, it needs to be brought well up to date, as its feels pretty creaky...

Sadly yes, I assume the "day job" where the owners earn money has taken more of their time than before.

Sure, but it will only grow in time. My point is they are behind the curveball here, when you can easily test up to 10 Gbps on off the shelf speedtest.net using various clients etc.

Always strikes me how well Mark does as a one man band over at ISPreview juggling the various demands of news reporting, forum etc

In reply to a post by jchamier:

I'm a fan of wget or curl a super large file and timing it, but that usually is only single connection so in modern WAN networks it appears you need multiple connections to show total throughput.

This is an artefact of the delay-bandwidth product and low level packet loss. The longer the round-trip time to a destination, the harder it is to achieve high TCP throughput.

https://en.wikipedia.org/wiki/TCP_tuning#Packet_loss

e.g. if your destination is 8ms away, and the packet loss is 0.01% (that is, only 1 packet in 10,000 is dropped), then the maximum throughput of a single TCP stream is

1460 / 0.008 / sqrt(0.0001) = 18.25MB/s = 146 Mb/s

This applies to each TCP stream individually, so running multiple concurrent TCP streams increases your total throughput.

Such low-level packet loss can indicate a problem in someone's network somewhere (e.g. dirt on a connector or buffers not sized for microbursts) - but getting your ISP to diagnose such an issue, especially through their upstream transit networks, is pretty much impossible. And it could be a problem at your end, of course.

This sort of optimization *does* happen in R&E networks though, where petabyte datasets are being thrown around. Tools like perfsonar are used to measure the packet loss at very high resolution (by default 10 packets per second = 36,000 packets per hour, measured separately in each direction)

Thank you fascinating analysis!

In reply to a post by Pheasant:

Always strikes me how well Mark does as a one man band over at ISPreview juggling the various demands of news reporting, forum etc

I think the forum here is quite well behaved, I'm not really up on the ISPreview forum, but Mark's news has taken over this place.

Yeah both pretty well behaved. You get the odd weirdo and oddball, but the same anywhere and in any forum I’ve ever used.

I’m glad ISPreview is kept reasonably fresh and up to date.

It’s a slightly different demographic over there too. My impression is a lot of older folks over here, especially the regulars. Perhaps it’s the familiarity of the interface, look and feel. Hasn’t changed in twenty something years. Nothing wrong with that and no disrespect. Just an observation.

In reply to a post by Pheasant:

Nothing wrong with that and no disrespect. Just an observation.

Its the only threaded forum I've used in years, without gimmicks such as "Kudos" that clutter up places such as the official ISP run forums; along with huge avatars. The downside is as Seb reported, its based on an old product with their own modifications. Seb & John have posted they have plans, but I guess time is not easy.