Regarding the speed test sites that we use; personally, I would prefer to use the Thinkbroadband Speed Tests to the others but perhaps that is because its the one that I have come to know and trust over the years. - The Fast,com speed test seems to be off and I do not know if the Ookla speed test is reliable but that does seem to be in the ball park of the Thinkbroadband speed tests;

https://zen-internet.speedtestcustom.com/result/312c...

and

https://www.thinkbroadband.com/speedtest/16025363266...

Someone suggested that I use a CAT7 cable but I seem to recall the Fritzbox Flat Cables are CAT7, (I have never had any issues with them), however, I did try a three meter CAT7 Cable from Amazon Basics that I had in my garage and I tried a new CAT6A cable, (that I made up myself for the test using a length of CAT6A cable and a couple of CAT7 plugs that were left over when we had the house wired for ethernet throughout), but that made no difference. - Once raised as an option; it did need to be considered because when I plugged the PC directly into the ONT, (prior to my making my first post on this thread), I had simply moved the ethernet cable from the router WAN Port to the PC RJ45 Port).

Someone asked if my equipment was capable of the speeds but I did point out in the earlier posts that it does sometimes run at over 900 Mbps , also my my PC and its ethernet card are rated to suit the speeds, (prior to my making my first post), so that is not an issue.- I use Windows 10, I keep it updated and I do not tweak it; I use Kaspersky Total with its tools to remove any rubbish. - I have tested the line on other equipment, including an almost new Xbox One X and the reported speeds on the Xbox One X vary in time with those on the PC.

Soon my circuit is to be given Elevated Traffic Status as that is what I pressed for as a first step because the general consensus seems to be that it is all down to contention and I am optimistic that it will cure the issues.

We do need it cured since, as can be seen from the above speed tests, today it is just more of the same.

Are you temporarily disabling Kaspersky Total Security when you do your speed tests? I use Kaspersky too, and found that it affected my speed test results.

Have you followed the earlier advise of booting from a Linux Live USB and doing the test there?

I mostly use Linux at home and have seen many occasions where Windows 10 achieves random throughput whereas Linux consistently hits my maximum combined speed. (not on FTTP mind you, dual FTTC)

In reply to a post by Fido:

Someone suggested that I use a CAT7 cable but I seem to recall the Fritzbox Flat Cables are CAT7, (I have never had any issues with them), however, I did try a three meter CAT7 Cable from Amazon Basics that I had in my garage and I tried a new CAT6A cable, (that I made up myself for the test using a length of CAT6A cable and a couple of CAT7 plugs that were left over when we had the house wired for ethernet throughout), but that made no difference. - Once raised as an option; it did need to be considered because when I plugged the PC directly into the ONT, (prior to my making my first post on this thread), I had simply moved the ethernet cable from the router WAN Port to the PC RJ45 Port).

You do know that Cat 7 cable is a complete and utter waste of time? In fact it is worse than that it is likely to negatively effect things unless you are using GG45/AR45 or TERA connectors with suitable sockets; which you 100% are not doing.

Further it is not recognized in any IEEE standard for Ethernet or is a TIA/EIA standard. 25Gbase-T and 40gbase-T both require Cat8 cable which is stated in the standard as being for data centre usage. Frankly given the incredibly low penetration of 10Gbase-T in the data centre (it's all direct attach SFP+ or actual fibre) then I imagine the take up of this will be even lower.

TL;DR anything beyond Cat 6a is only used by the uniformed and gullible who have more money than sense. If you genuinely have a noise problem use FTP Cat6a instead of UTP.

My download speeds were particularly poor with a PPPoE connection on my Windows 10 laptop, which is why I suggested that the OP used a bootable Linux USB for their testing.

In reply to a post by jabuzzard:

You do know that Cat 7 cable is a complete and utter waste of time?

I would like to say that I did know that but I did not.

That said; I don't tend to use CAT7, (which is why I had to search the garage for an old CAT7e cable).

I do usually use CAT6A for more critical areas and also for the fixed infrastructure but for most of the house I use whatever cables come with the equipment for TV's etc. and these are probably all CAT5e.

Regarding Elevated Traffic Status; apparently they put me on this today and I am not sure if does a lot of good since it has been up and down since the start. - One thing that I have noticed is that the Xbox Latency has gone right up and that world be a real negative if that was to continue. - (I am not sure why, perhaps this Elevated Traffic Status is a form of interleaving. - I don't know).

I also pressed the reset button on the ONT but I do not know if that had any effect.

Regarding a Linux USB: today, I did create a Linux USB; it seems to be something called Mint/main cinamon Linux and that does seem to operate in a fashion but it's hard to say as the speeds on Windows 10 have been up and down for a week.

https://www.thinkbroadband.com/speedtest/16026278694...

and

https://www.thinkbroadband.com/speedtest/16026282797...

In some ways using a Linux USB is a bit like taking all of the seats, all the doors, the roof and almost everything else off a passenger car to see if you can go a few miles faster in the wind and the rain which is not a real world solution and it's not an answer but it could be useful as a comparison .

I will see how it goes over the next week.

In reply to a post by Fido:

In some ways using a Linux USB is a bit like taking all of the seats, all the doors, the roof and almost everything else off a passenger car to see if you can go a few miles faster in the wind and the rain which is not a real world solution and it's not an answer but it could be useful as a comparison .

I will see how it goes over the next week.

Have the hood down on a car and no chance to close it when it starts raining, you need to go fast! That way the airflow over the windscreen carries nearly all of it away, except in a downpour.

This is what Zen told me when they enabled elevated traffic status for me:

At the moment, the resolution we've used on your service (and on some others) is to request an additional feature for your service called elevated traffic status. This prioritises your network traffic above others (standard traffic) meaning if there's any congestion, you should be unaffected.

As you can see from my BQM Graph, I have had no problems with latency since elevated traffic status was enabled.

This test https://www.thinkbroadband.com/speedtest/16026282797... suggests nothing at all wrong with the FTTP or ISP side of things.

As this was Linux as opposed to Windows it points towards an issue with your Windows setup, network drivers, anti virus, previous speed tuning maybe.

Only way to more accurately test that is:

1. Boot the Linux USB - do the Test - note results
2. Boot back to Windows - do the test - note results

I know other variables go on in the minutes between tests (server load / ISP load) but should give a good idea whether it is Windows. It would NOT be unusual for Win10 (fully updated) to affect Network traffic, we had a similar issue on our enterprise network not long ago.

@Fido - You need to consistently test this at varying times per day (doing the linux test, then the windows test in relatively quick successfion).

Once this is done for a significant period of time (a week maybe) you will have some relatively useful data to go on to take to your ISP or investigate your own equipment.

Bear in mind XBOX's use Windows 10 also with a different UI on them so may not be a good test either