TBH , the car analogy could be presented differently, it’s like complaining that although the car you were sold is supposedly capable of 0-60 in under 5 seconds and 200MPH , you cannot get near those performance figures because of the other traffic on the road ( ignoring the legal issues with those speeds ) , and additionally, your BHP analogy, if the company you employed to confirm the engine horsepower, didn’t use the appropriate ‘kit’ to confirm the manufacturer claims , and all they did was take the car for a spin around the block , and base their opinion on that journey , the manufacturer would ( quite rightly) say that ‘test’ was of no value .
In the OP case , they confirm they get the speed they pay for on some equipment/setup and not on another , that pretty much indicates where the issue is.
In your case , provided the profile allocated to you is correct , and the only profile that gets the 100Mb upload is the 900Mb , so that suggests yours is correct, try at a likely quiet time and if the speed is better , perhaps it’s congestion on your PON ( you share 2.4Gb download with others , perhaps they are hammering their connections at the same time you are running speed your tests ) , if there is no difference between quiet times and busy times ( generally taken to be 6-9pm ) then it’s not likely to be congestion.
Fttp doesn’t have any rate adaption, you sync at the speed rate you pay for ( with the proviso that that has been correctly set by the administrator) throughput is not under your provider control, and WiFi isn’t suitable for any measure of speed , also some people will ‘tests’ although using a wired connection, but using kit that is not Gb capable or incorrectly set up.
If the router ( SH2 ) is suspect, connect a suitable PC directly to the ONT , set up a PPPoE session with a suitable wired device and run the speed test from there.
